- Email: [email protected]
Elevated glycated hemoglobin level and hyperglycemia after aneurysmal subarachnoid hemorrhage
Accepted Manuscript Title: Elevated glycated hemoglobin level and hyperglycemia after aneurysmal subarachnoid hemorrhage Authors: Kerim Beseoglu, Hans-Jakob Steiger PII: DOI: Reference:
S0303-8467(17)30304-9 https://doi.org/10.1016/j.clineuro.2017.10.037 CLINEU 4822
To appear in:
Clinical Neurology and Neurosurgery
Received date: Revised date: Accepted date:
23-8-2017 17-10-2017 29-10-2017
Please cite this article as: Beseoglu Kerim, Steiger Hans-Jakob.Elevated glycated hemoglobin level and hyperglycemia after aneurysmal subarachnoid hemorrhage.Clinical Neurology and Neurosurgery https://doi.org/10.1016/j.clineuro.2017.10.037 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Elevated glycated hemoglobin level and hyperglycemia after aneurysmal subarachnoid hemorrhage Kerim Beseoglu M.D.1, Hans-Jakob Steiger M.D.1, 1 Department of Neurosurgery, Medical Faculty, Heinrich-Heine-University, Düsseldorf, Germany Corresponding Author Kerim Beseoglu Department of Neurosurgery University Clinic Düsseldorf Moorenstrasse 5 40225 Düsseldorf, Germany Tel: 0049-211-81-07329 Fax: 0049-211-81-18818 Email: [email protected]
Highlights - Elevated blood glucose correlates with poor outcome in subarachnoid hemorrhage - Initial hyperglycemia correlates with poor neurological condition - A pre-existing hyperglycemic metabolism does not contribute to poor outcome - Elevated HbA1c does not increase risk for delayed cerebral ischemia
1
Abstract Objectives. Elevated blood glucose is frequently detected early after aneurysmal subarachnoid hemorrhage (aSAH) and is considered a risk factor for poor neurological outcome. However it remains unclear whether hyperglycemia is caused by the SAH ictus or reflects a pre-existing hyperglycemic metabolism. In a prospective register we analysed glycated haemoglobin levels (HbA1c) in patients with aSAH and its influence on outcome. Patients and Methods. Between July 2012 and July 2014, 87 patients with confirmed aSAH were included (NCT02081820). Within 72 hours HbA1c levels were assessed as a measure for hyperglycemic metabolism preceding aSAH. Blood glucose levels were recorded upon admission. Patient outcome was recorded after 6 months using modified Rankin scale (mRS). Results. HbA1c levels did not correlate with initial neurological status (p=0.338, r=0.104). On the contrary, initial blood glucose levels correlated significantly with neurological status at admission (p=0.001, r=0.341). Additionally, HbA1c levels failed to show a significant influence on the occurrence of delayed cerebral ischemia (DCI) (p=0.400) or outcome after 6 months (p=0.790). Conclusion. A pre-existing hyperglycemic metabolism does not contribute to the severity of aSAH or influences the quality of neurological recovery. Hyperglycemia after aSAH correlates with initial neurological status and patient outcome and is potentially attributable to the metabolic changes induced by the brain injury after the hemorrhage.
Key words:
aneurysm, HbA1c, Hyperglycemia, outcome, subarachnoid hemorrhage 2
Introduction Hyperglycemia is a common phenomenon in the acute phase after aneurysmal subarachnoid hemorrhage (aSAH) and is linked to an increased risk of poor outcome [1]. Multiple mechanisms causing blood glucose elevation are currently discussed. Activation of the sympathetic autonomic nervous system by subarachnoid blood and the subsequent increase in catecholamine and cortisol levels induces hyperglycemia and insulin resistance. Concomitantly, an inflammatory response is triggered leading to the release of cytokines, which in turn aggravate the stress response and is associated with hyperglycemia and insulin resistance. Additionally, direct injury to hypothalamic nuclei occurs in a great number of patients after aSAH, thus an alteration of the hypothalamicpituitary-adrenal axis impairing glucose homeostasis is imaginable [2]. The influence of abnormalities of glucose metabolism on the incidence aSAH is considered to be of minor relevance, especially since neither lack of insulin nor insulin resistance are a risk factor for aneurysm rupture [3]. However, robust data on the influence of a preexistent dysfunction in glucose metabolism is rare and therefore its role remains uncertain. Here, we analyze whether a preexistent hyperglycemia as reflected in an elevated fraction of glycated hemoglobin (HbA1c) contributes to patient outcome and initial blood glucose level in patients after aSAH.
3
Material and Methods We prospectively included all patients with the admission diagnosis subarachnoid hemorrhage between July 2012 and July 2014 in a single center patient registry (Clinical Trial Registration: http://www.clinicaltrials.gov. Unique identifier: NCT02081820). The registry was approved by the ethics committee of the Heinrich-Heine University Düsseldorf (study number: 4184) Inclusion criteria for this analysis were (1) subarachnoid hemorrhage from a ruptured cerebral artery aneurysm confirmed by either computed tomography or digital subtraction angiogram, (2) age over 18 years, (3) admission within 72 hours after aSAH and (4) survival ≥ 5 days after hemorrhage. None of the included patients had a history of diabetes or received oral antidiabetic medication or insulin. Patients with non-aneurysmal SAH or other underlying pathologies causing the SAH (e.g. arterio-venous malformation, cavernoma) were excluded. The management of included patients was in accordance with current SAH guidelines [4]. Epidemiological data such as age, sex, clinical and neurological status at admission and amount of subarachnoid blood on computerized tomography (CT) scan were recorded. Initial neurological status was graded according to World Federation of Neurosurgical Societies (WFNS) grade and grouped into good grade (WFNS grade 1-3) and poor grade patients (WFNS grade 4-5) [5]. The subarachnoid blood distribution was recorded according to the modified Fisher scale [6]. The first non-fasting blood glucose level (BG) in mmol/L assessed after admission was defined as admission BG and BG > 7.8mmol/L was defined as hyperglycemia [7]. Blood samples were taken before any glucose lowering therapy was initiated and none of the patients received pro4
glycemic medication, e.g. steroids. By exposure to blood glucose, a fraction of hemoglobin corresponding to the plasma glucose level is non-enzymatically glycated. This stable glycated hemoglobin form (HbA1c) serves as a reliable marker for the average blood glucose level of up to three months preceding the measurement [8]. According to international standards HbA1c values were recorded in International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) units [mmol/mol] [9]. According to recommendations of the American Diabetes Association and the U.S. National Institute of Diabetes and Digestive and Kidney Diseases values above 39 mmol/mol were considered pathological for non-diabetic adults [10]. The primary end-point was functional outcome as defined by the mRS at 6 months after aSAH. Additionally, incidence of delayed cerebral ischemia (DCI) and new cerebral infarction attributable to DCI were recorded as secondary end-points. DCI was defined as a new neurological deficit or a reduction in vigilance of at least 2 points on the Glasgow Coma Scale not attributable to other causes by means of clinical or radiological assessment and new cerebral infarction from DCI was defined as the presence of infarction on radiographic imaging at the time of discharge not present on imaging after aneurysm occlusion and not attributable to treatment procedure or cerebral hematoma as previously published [11]. In addition to clinical monitoring we correlated perfusion computerized tomography (PCT) results (mean transit time, MTT) obtained immediately after admission and prior to aneurysm occlusion (early MTT, eMTT) and peak MTT (pMTT) defined as the maximum MTT within 14 days after SAH with initial BG and HbA1c level. PCT measurement and analysis was performed as previously described by us [12]. Statistical analysis included correlation analysis of HbA1c and initial glucose level with admission neurological status and its relation with outcome and calculation of the relative risk
5
(RR) for reduced neurological admission status, incidence of DCI and worse outcome using SPSS 15 software (SPSS Inc., Chicago, IL, USA). Significance was accepted when p<0.05.
Results Into the registry 168 patients were included. The flow chart in figure 1 illustrates the process of patient inclusion and exclusion for this analysis (Figure 1). Eighty-seven patients with complete data sets were available for analysis. Included were 57 female and 30 male patients (mean age 56.9 ±10.9 years). Relevant epidemiological data are summarized in table 1 and table 2 (Table 1 and Table 2). Hyperglycemia was present in 40 (45.5%) patients, whereas pathological HbA1c was detected in only 26 (29.5%) patients. Detailed laboratory values on HbA1c and BG levels are given in table 3 (Table 3). HbA1c level did not correlate with initial neurological status (p=0.338, r=0.104) or the amount of subarachnoid blood (modified Fisher grade, p=0.190, r=0.142). Pathological HbA1c did not increase the risk for poorer neurological admission status (RR 1.466 95%CI 0.932-2.307, p=0.098). On the contrary, initial blood glucose level correlated significantly with neurological status at admission (p=0.001, r=0.341) and modified Fisher score (p<0.001, r=0.484). HbA1c failed to show a significant correlation with the occurrence of DCI (p=0.400, r=-0.091) or DCI-related infarction (p=0.268, r=0.120). The risk for the occurrence of DCI was not elevated in patients with HbA1c above 39mmol/l (RR 0.902 95%CI 0.512-1.590, p=0.355) or in patients with initial hyperglycemia (RR 0.857 95%CI 0.513-1.431, p=0.556).
6
Early MTT did not correlate with BG (p=0.341, r=0.132) or HbA1c (p=0.661, r=0.061) elevation. While elevated pMTT >4.2s correlated with the occurrence DCI-related infarction (p=0.025, r=0.225), it showed no correlation with HbA1c or BG (p=0.699, r=0.043 and p=0.706, r=0.042, respectively). HbA1c failed to show a significant correlation with patient outcome (mRS) after 6 months (p=0.790, r=0.029). In contrast, initial BG correlated with patient outcome after 6 months (p=0.004, r=0.302). Risk for poorer long-term outcome (RR 0.816 95% CI 0.421-1.580, p=0.547) was not increased in patients with elevated HbA1c. However, in patients presenting with hyperglycemia the relative risk for poor outcome (mRS 3-6) increased by 2.19 (p=0.014, 95% CI 1.17-4.11). Despite these correlations, a binary logistic regression model could not show a significant contribution of BG to neurological outcome when compared with known confounders such as WFNS grade and modified Fisher score.
Discussion In this analysis we can demonstrate, that first, elevated HbA1c in patients with aSAH does not correlate with neurological status at admission in contrast to elevated BG. Furthermore, HbA1c is not a relevant predictor for the occurrence of DCI, DCI-related infarction or neurological outcome. Lastly, independent of HbA1c, elevated initial BG serves as a significant predictor of poor outcome after 6 months. Diabetes is associated with an increased frequency and severity of ischemic stroke and hyperglycemia early after stroke onset appears to be an independent contributor to poorer outcome [13]. Comparably, admission hyperglycemia in aSAH is associated with poorer 7
outcome, however, a relevant increase in risk of aneurysm rupture and poorer outcome in diabetic patients could not be demonstrated [1, 3]. The proposed pathophysiological explanation for hyperglycemia attributes this phenomenon to the direct brain injury induced by the hemorrhage through various direct and indirect pathways [2]. Our data illustrates a dissociation of HbA1c and BG after aSAH with a disproportionally higher number of hyperglycemic patients compared to a lower number of patients with pathological HbA1c. Consequently, initial BG appears to correlate more with the severity of the hemorrhage reflected by poorer neurological condition and thus, a possible explanation could be that hyperglycemia may be induced by the acute injury to the brain independent of the pre-ictal degree of glycemia. Though the definition of hyperglycemia remains inconsistent between various studies, elevated BG impacts neurological outcome [1]. In a microdialysis evaluation of cerebral glucose metabolism a threshold for serum BG of 7.5 mmol/l could best distinguish between good and poor outcome and BG > 7.8 mmol/l independently predicted poor outcome [7]. Additionally, occurrence of stress-induced hyperglycemia (a ratio of admission glucose levels and an estimation of preadmission glucose levels) appears to predict necessity for placement of external ventricular drainage [14]. Neither admission blood glucose nor history of diabetes mellitus did increase the risk for symptomatic vasospasm in a retrospective analysis of 352 patients and rates of DCI were comparable between patients with stress-induced hyperglycemia and those without in another study [14, 15]. Concordantly, cerebral perfusion as monitored by early PCT and incidence of DCI were not directly affected by initial hyperglycemia or elevated HbA1c in our patients. Initial hyperglycemia may thus be attributed to metabolic changes caused by early brain injury. However, this relates only to the immediate situation after the hemorrhage, as hyperglycemia in 8
the days following aSAH may increase risk of DCI and aggravates cerebral vasospasm in animal experiment [7, 16, 17]. Limitations This short report holds several limitations. The patient cohort is small and from a single center hindering generalizability. A lack of blinding towards the primary and secondary endpoints exists and a high number of potentially eligible patients had incomplete data sets and were excluded. This leads to a distortion of the patient cohort and contributes to a potentially relevant type-2 error. Moribund patients were excluded from the analysis because in these cases diagnostic and/or therapeutic procedures were often withheld due the very unfavorable prognosis leading to inconsistent data. However, poor grade patients offer interesting insights in the pathophysiology of early brain injury and excluding these patients constitutes a limitation of this analysis. Not all contributing variables known to influence outcome after aSAH were included in our analysis. Especially, modifiable variables are susceptible for therapeutic interventions and exclusion of these variables (e. g. initial blood pressure and use of anticonvulsants) may distort our analysis where neurological outcome was chosen as primary end-point. However, in multivariable logistic and Cox proportional hazards regression initial neurological grade, subarachnoid blood volume and patient age are still among the most relevant contributors and these were included in our analysis [18]. A non-significant trend for higher patient age and poorer neurological grade at admission in patients with elevated HbA1c possibly indicates higher relevance of HbA1c, but this might be 9
obscured by the small cohort size. Additionally, a relevant disproportion in the treatment modality between low and high HbA1c patient groups may have affected end-points.
Conclusion A hyperglycemic metabolism preceding aneurysm rupture, as reflected in an elevated HbA1c, does not contribute to poorer neurological outcome. Hyperglycemia after aSAH correlates with initial neurological status and patient outcome and is potentially attributable to the metabolic changes induced by the brain injury after the hemorrhage.
Acknowledgements None
All listed authors contributed in conceiving the study, analyzing data and preparing and drafting the manuscript. The authors state no conflict of interest and have nothing to disclose. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
10
References 1.
Kruyt, N.D., G.J. Biessels, R.J. de Haan, M. Vermeulen, G.J. Rinkel, B. Coert, et al., Hyperglycemia and clinical outcome in aneurysmal subarachnoid hemorrhage: a metaanalysis. Stroke 2009;40(6):e424-30.
2.
Kruyt, N.D., G.J. Biessels, J.H. DeVries, M.J. Luitse, M. Vermeulen, G.J. Rinkel, et al., Hyperglycemia in aneurysmal subarachnoid hemorrhage: a potentially modifiable risk factor for poor outcome. J Cereb Blood Flow Metab 2010;30(9):1577-87.
3.
Lindgren, A.E., M.I. Kurki, A. Riihinen, T. Koivisto, A. Ronkainen, J. Rinne, et al., Type 2 diabetes and risk of rupture of saccular intracranial aneurysm in eastern Finland. Diabetes Care 2013;36(7):2020-6.
4.
Connolly, E.S., Jr., A.A. Rabinstein, J.R. Carhuapoma, C.P. Derdeyn, J. Dion, R.T. Higashida, et al., Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/american Stroke Association. Stroke 2012;43(6):1711-37.
5.
Teasdale, G.M., C.G. Drake, W. Hunt, N. Kassell, K. Sano, B. Pertuiset, et al., A universal subarachnoid hemorrhage scale: report of a committee of the World Federation of Neurosurgical Societies. J Neurol Neurosurg Psychiatry 1988;51(11):1457.
6.
Frontera, J.A., J. Claassen, J.M. Schmidt, K.E. Wartenberg, R. Temes, E.S. Connolly, Jr., et al., Prediction of symptomatic vasospasm after subarachnoid hemorrhage: the modified fisher scale. Neurosurgery 2006;59(1):21-7; discussion 21-7.
11
7.
Schlenk, F., P. Vajkoczy, A. Sarrafzadeh, Inpatient hyperglycemia following aneurysmal subarachnoid hemorrhage: relation to cerebral metabolism and outcome. Neurocrit Care 2009;11(1):56-63.
8.
Nathan, D.M., H. Turgeon, S. Regan, Relationship between glycated haemoglobin levels and mean glucose levels over time. Diabetologia 2007;50(11):2239-44.
9.
Consensus Statement on the Worldwide Standardization of the Hemoglobin A1C Measurement: The American Diabetes Association, European Association for the Study of Diabetes, International Federation of Clinical Chemistry and Laboratory Medicine, and the International Diabetes Federation. Diabetes Care 2007;30(9):2399-2400.
10.
2. Classification and Diagnosis of Diabetes. Diabetes Care 2016;39 Suppl 1:S13-22.
11.
Vergouwen, M.D., M. Vermeulen, J. van Gijn, G.J. Rinkel, E.F. Wijdicks, J.P. Muizelaar, et al., Definition of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage as an outcome event in clinical trials and observational studies: proposal of a multidisciplinary research group. Stroke 2010;41(10):2391-5.
12.
Etminan, N., K. Beseoglu, H.J. Heiroth, B. Turowski, H.J. Steiger, D. Hanggi, Early perfusion computerized tomography imaging as a radiographic surrogate for delayed cerebral ischemia and functional outcome after subarachnoid hemorrhage. Stroke 2013;44(5):1260-6.
13.
Baird, T.A., M.W. Parsons, P.A. Barber, K.S. Butcher, P.M. Desmond, B.M. Tress, et al., The influence of diabetes mellitus and hyperglycaemia on stroke incidence and outcome. J Clin Neurosci 2002;9(6):618-26. 12
14.
Ray, B., A. Ludwig, L.K. Yearout, D.M. Thompson, B.N. Bohnstedt, Stress-Induced Hyperglycemia After Spontaneous Subarachnoid Hemorrhage and Its Role in Predicting Cerebrospinal Fluid Diversion. World Neurosurg 2017;100:208-215.
15.
Badjatia, N., M.A. Topcuoglu, F.S. Buonanno, E.E. Smith, R.G. Nogueira, G.A. Rordorf, et al., Relationship between hyperglycemia and symptomatic vasospasm after subarachnoid hemorrhage. Crit Care Med 2005;33(7):1603-9; quiz 1623.
16.
Naidech, A.M., K. Levasseur, S. Liebling, R.K. Garg, M. Shapiro, M.L. Ault, et al., Moderate Hypoglycemia is associated with vasospasm, cerebral infarction, and 3-month disability after subarachnoid hemorrhage. Neurocrit Care 2010;12(2):181-7.
17.
Huang, Y.H., C.L. Chung, H.P. Tsai, S.C. Wu, C.Z. Chang, C.Y. Chai, et al., Hyperglycemia Aggravates Cerebral Vasospasm after Subarachnoid Hemorrhage in a Rat Model. Neurosurgery 2017;80(5):809-815.
18.
Rosengart, A.J., K.E. Schultheiss, J. Tolentino, R.L. Macdonald, Prognostic factors for outcome in patients with aneurysmal subarachnoid hemorrhage. Stroke 2007;38(8):231521.
13
Figure legend Figure 1. Flow chart of the patient inclusion and exclusion process
14
Table legends Table 1. Epidemiological data of the included 87 patients stratified into normal and elevated HbA1c levels with test results for group differences. BG denotes blood glucose, WFNS World Federation of Neurological Surgeons Grade, DCI delayed cerebral infarction. Percentages are rounded. Table 2. Epidemiological data of the included 87 patients stratified into normal and elevated blood glucose levels with test results for group differences. BG denotes blood glucose, WFNS World Federation of Neurological Surgeons Grade, DCI delayed cerebral infarction. Percentages are rounded. Table 3. Mean values for HbA1c and blood glucose (BG) levels with standard deviation in brackets and range for all patients and grouped for patients with normal versus elevated HbA1c levels. Group difference was tested using the Student’s t-test corrected for unequal variance.
15
Table 1. Epidemiological data of the included 87 patients stratified into normal and elevated HbA1c levels with test results for group differences. BG denotes blood glucose, WFNS World Federation of Neurological Surgeons Grade, DCI delayed cerebral infarction. Percentages are rounded.
N (%)
All
HbA1c <40mmol/mol
HbA1c >39mmol/mol
87 (100)
61 (70)
26 (30)
56.9 (10.9)
55.5 (10.4)
60.9 (10.9)
0.066 0.630
Age
Years (SD)
Sex
F (%)
57 (66)
41 (67)
16 (62)
M (%)
30 (34)
20 (33)
10 (38)
<7.9mmol/L (%)
47 (54)
33 (54)
14 (54)
>7.8mmol/L (%)
40 (46)
28 (46)
12 (46)
1-3 (%)
48 (55)
37 (61)
11 (42)
4-5 (%)
39 (45)
24 (39)
15 (58)
0 (%)
3 (3)
2 (3)
0
1 (%)
18 (21)
13 (21)
5 (19)
2 (%)
6 (7)
6 (10)
0
3 (%)
45 (51)
30 (49)
15 (58)
4 (%)
16 (18)
10 (16)
6 (23)
Anterior circulation (%)
63 (73)
41 (73)
22 (85)
Posterior circulation (%)
19 (22)
15 (27)
4 (15)
BG
WFNS
Fisher
Aneurysm
Other (%)
5 (5)
16
p
0.983
0.115
0.206
0.461
Treatment
Surgery (%)
45 (52)
36 (61)
9 (35)
Endovascular (%)
40 (46)
23 (39)
17 (65)
None (%) DCI
Infarction
Outcome
0.050
2 (2)
No (%)
51 (59)
35 (57)
16 (62)
Yes (%)
36 (41)
26 (43)
10 (38)
No (%)
73 (84)
51 (84)
22 (85)
Yes (%)
14 (16)
10 (16)
4 (15)
Good (mRS 0-2)
56 (64)
38 (62)
18 (69)
Poor (mRS 3-6)
31 (36)
23 (38)
8 (31)
17
0.718
0.907
0.536
Table 2. Epidemiological data of the included 87 patients stratified into normal and elevated blood glucose levels with test results for group differences. BG denotes blood glucose, WFNS World Federation of Neurological Surgeons Grade, DCI delayed cerebral infarction. Percentages are rounded.
N (%)
All
BG < 7.9 mmol/l
BG > 7.8 mmol/l
87 (100)
47
40
56.9 (10.9)
55.3 (11.2)
59.3 (10.0)
0.034 0.009
Age
Years (SD)
Sex
F (%)
57 (66)
25(44)
32(56)
M (%)
30 (34)
22(73)
8(27)
<40mmol/mol (%)
47 (54)
33(54)
28(46)
>39mmol/mol (%)
40 (46)
14(54)
12(46)
1-3 (%)
48 (55)
32(67)
16(33)
4-5 (%)
39 (45)
15(39)
24(61)
0 (%)
3 (3)
2(100)
0
1 (%)
18 (21)
14(78)
4(22)
2 (%)
6 (7)
3(50)
3(50)
3 (%)
45 (51)
24(53)
21(47)
4 (%)
16 (18)
4(25)
12(75)
Anterior circulation (%)
63 (73)
39(61)
25(39)
Posterior circulation (%)
19 (22)
4(24)
13(76)
HbA1c
WFNS
Fisher
Aneurysm
Other (%)
5 (5)
18
p
1.000
0.010
0.002
0.012
Treatment
Surgery (%)
45 (52)
26(58)
19(42)
Endovascular (%)
40 (46)
19(48)
21(52)
2 (2)
2(100)
0
No (%)
51 (59)
26(51)
25(49)
Yes (%)
36 (41)
21(58)
15(42)
No (%)
73 (84)
41(56)
32(44)
Yes (%)
14 (16)
6(43)
8(57)
Good (mRS 0-2)
56 (64)
36(64)
20(36)
Poor (mRS 3-6)
31 (36)
11(36)
20(64)
None (%) DCI
Infarction
Outcome
0.609
0.521
0.394
0.014
Table 3. Mean values for HbA1c and blood glucose (BG) levels with standard deviation in brackets and range for all patients and grouped for patients with normal versus elevated HbA1c levels. Group difference was tested using the Student’s t-test corrected for unequal variance.
All patients
HbA1c <40mmol/mol
HbA1c >39mmol/mol
87
61
26
HbA1c [mmol/mol]
38.6(±9.9) 25-107
34.7(±2.9) 25-39
47.7(±14.0) 40-107
<0.01
BG [mmol/l]
8.5(±2.5) 4.7-16.7
8.2(±2.1) 4.7-15.4
9.3(±3.1) 5.8-16.7
0.103
N=
19
P=
20
S0303-8467(17)30304-9 https://doi.org/10.1016/j.clineuro.2017.10.037 CLINEU 4822
To appear in:
Clinical Neurology and Neurosurgery
Received date: Revised date: Accepted date:
23-8-2017 17-10-2017 29-10-2017
Please cite this article as: Beseoglu Kerim, Steiger Hans-Jakob.Elevated glycated hemoglobin level and hyperglycemia after aneurysmal subarachnoid hemorrhage.Clinical Neurology and Neurosurgery https://doi.org/10.1016/j.clineuro.2017.10.037 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Elevated glycated hemoglobin level and hyperglycemia after aneurysmal subarachnoid hemorrhage Kerim Beseoglu M.D.1, Hans-Jakob Steiger M.D.1, 1 Department of Neurosurgery, Medical Faculty, Heinrich-Heine-University, Düsseldorf, Germany Corresponding Author Kerim Beseoglu Department of Neurosurgery University Clinic Düsseldorf Moorenstrasse 5 40225 Düsseldorf, Germany Tel: 0049-211-81-07329 Fax: 0049-211-81-18818 Email: [email protected]
Highlights - Elevated blood glucose correlates with poor outcome in subarachnoid hemorrhage - Initial hyperglycemia correlates with poor neurological condition - A pre-existing hyperglycemic metabolism does not contribute to poor outcome - Elevated HbA1c does not increase risk for delayed cerebral ischemia
1
Abstract Objectives. Elevated blood glucose is frequently detected early after aneurysmal subarachnoid hemorrhage (aSAH) and is considered a risk factor for poor neurological outcome. However it remains unclear whether hyperglycemia is caused by the SAH ictus or reflects a pre-existing hyperglycemic metabolism. In a prospective register we analysed glycated haemoglobin levels (HbA1c) in patients with aSAH and its influence on outcome. Patients and Methods. Between July 2012 and July 2014, 87 patients with confirmed aSAH were included (NCT02081820). Within 72 hours HbA1c levels were assessed as a measure for hyperglycemic metabolism preceding aSAH. Blood glucose levels were recorded upon admission. Patient outcome was recorded after 6 months using modified Rankin scale (mRS). Results. HbA1c levels did not correlate with initial neurological status (p=0.338, r=0.104). On the contrary, initial blood glucose levels correlated significantly with neurological status at admission (p=0.001, r=0.341). Additionally, HbA1c levels failed to show a significant influence on the occurrence of delayed cerebral ischemia (DCI) (p=0.400) or outcome after 6 months (p=0.790). Conclusion. A pre-existing hyperglycemic metabolism does not contribute to the severity of aSAH or influences the quality of neurological recovery. Hyperglycemia after aSAH correlates with initial neurological status and patient outcome and is potentially attributable to the metabolic changes induced by the brain injury after the hemorrhage.
Key words:
aneurysm, HbA1c, Hyperglycemia, outcome, subarachnoid hemorrhage 2
Introduction Hyperglycemia is a common phenomenon in the acute phase after aneurysmal subarachnoid hemorrhage (aSAH) and is linked to an increased risk of poor outcome [1]. Multiple mechanisms causing blood glucose elevation are currently discussed. Activation of the sympathetic autonomic nervous system by subarachnoid blood and the subsequent increase in catecholamine and cortisol levels induces hyperglycemia and insulin resistance. Concomitantly, an inflammatory response is triggered leading to the release of cytokines, which in turn aggravate the stress response and is associated with hyperglycemia and insulin resistance. Additionally, direct injury to hypothalamic nuclei occurs in a great number of patients after aSAH, thus an alteration of the hypothalamicpituitary-adrenal axis impairing glucose homeostasis is imaginable [2]. The influence of abnormalities of glucose metabolism on the incidence aSAH is considered to be of minor relevance, especially since neither lack of insulin nor insulin resistance are a risk factor for aneurysm rupture [3]. However, robust data on the influence of a preexistent dysfunction in glucose metabolism is rare and therefore its role remains uncertain. Here, we analyze whether a preexistent hyperglycemia as reflected in an elevated fraction of glycated hemoglobin (HbA1c) contributes to patient outcome and initial blood glucose level in patients after aSAH.
3
Material and Methods We prospectively included all patients with the admission diagnosis subarachnoid hemorrhage between July 2012 and July 2014 in a single center patient registry (Clinical Trial Registration: http://www.clinicaltrials.gov. Unique identifier: NCT02081820). The registry was approved by the ethics committee of the Heinrich-Heine University Düsseldorf (study number: 4184) Inclusion criteria for this analysis were (1) subarachnoid hemorrhage from a ruptured cerebral artery aneurysm confirmed by either computed tomography or digital subtraction angiogram, (2) age over 18 years, (3) admission within 72 hours after aSAH and (4) survival ≥ 5 days after hemorrhage. None of the included patients had a history of diabetes or received oral antidiabetic medication or insulin. Patients with non-aneurysmal SAH or other underlying pathologies causing the SAH (e.g. arterio-venous malformation, cavernoma) were excluded. The management of included patients was in accordance with current SAH guidelines [4]. Epidemiological data such as age, sex, clinical and neurological status at admission and amount of subarachnoid blood on computerized tomography (CT) scan were recorded. Initial neurological status was graded according to World Federation of Neurosurgical Societies (WFNS) grade and grouped into good grade (WFNS grade 1-3) and poor grade patients (WFNS grade 4-5) [5]. The subarachnoid blood distribution was recorded according to the modified Fisher scale [6]. The first non-fasting blood glucose level (BG) in mmol/L assessed after admission was defined as admission BG and BG > 7.8mmol/L was defined as hyperglycemia [7]. Blood samples were taken before any glucose lowering therapy was initiated and none of the patients received pro4
glycemic medication, e.g. steroids. By exposure to blood glucose, a fraction of hemoglobin corresponding to the plasma glucose level is non-enzymatically glycated. This stable glycated hemoglobin form (HbA1c) serves as a reliable marker for the average blood glucose level of up to three months preceding the measurement [8]. According to international standards HbA1c values were recorded in International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) units [mmol/mol] [9]. According to recommendations of the American Diabetes Association and the U.S. National Institute of Diabetes and Digestive and Kidney Diseases values above 39 mmol/mol were considered pathological for non-diabetic adults [10]. The primary end-point was functional outcome as defined by the mRS at 6 months after aSAH. Additionally, incidence of delayed cerebral ischemia (DCI) and new cerebral infarction attributable to DCI were recorded as secondary end-points. DCI was defined as a new neurological deficit or a reduction in vigilance of at least 2 points on the Glasgow Coma Scale not attributable to other causes by means of clinical or radiological assessment and new cerebral infarction from DCI was defined as the presence of infarction on radiographic imaging at the time of discharge not present on imaging after aneurysm occlusion and not attributable to treatment procedure or cerebral hematoma as previously published [11]. In addition to clinical monitoring we correlated perfusion computerized tomography (PCT) results (mean transit time, MTT) obtained immediately after admission and prior to aneurysm occlusion (early MTT, eMTT) and peak MTT (pMTT) defined as the maximum MTT within 14 days after SAH with initial BG and HbA1c level. PCT measurement and analysis was performed as previously described by us [12]. Statistical analysis included correlation analysis of HbA1c and initial glucose level with admission neurological status and its relation with outcome and calculation of the relative risk
5
(RR) for reduced neurological admission status, incidence of DCI and worse outcome using SPSS 15 software (SPSS Inc., Chicago, IL, USA). Significance was accepted when p<0.05.
Results Into the registry 168 patients were included. The flow chart in figure 1 illustrates the process of patient inclusion and exclusion for this analysis (Figure 1). Eighty-seven patients with complete data sets were available for analysis. Included were 57 female and 30 male patients (mean age 56.9 ±10.9 years). Relevant epidemiological data are summarized in table 1 and table 2 (Table 1 and Table 2). Hyperglycemia was present in 40 (45.5%) patients, whereas pathological HbA1c was detected in only 26 (29.5%) patients. Detailed laboratory values on HbA1c and BG levels are given in table 3 (Table 3). HbA1c level did not correlate with initial neurological status (p=0.338, r=0.104) or the amount of subarachnoid blood (modified Fisher grade, p=0.190, r=0.142). Pathological HbA1c did not increase the risk for poorer neurological admission status (RR 1.466 95%CI 0.932-2.307, p=0.098). On the contrary, initial blood glucose level correlated significantly with neurological status at admission (p=0.001, r=0.341) and modified Fisher score (p<0.001, r=0.484). HbA1c failed to show a significant correlation with the occurrence of DCI (p=0.400, r=-0.091) or DCI-related infarction (p=0.268, r=0.120). The risk for the occurrence of DCI was not elevated in patients with HbA1c above 39mmol/l (RR 0.902 95%CI 0.512-1.590, p=0.355) or in patients with initial hyperglycemia (RR 0.857 95%CI 0.513-1.431, p=0.556).
6
Early MTT did not correlate with BG (p=0.341, r=0.132) or HbA1c (p=0.661, r=0.061) elevation. While elevated pMTT >4.2s correlated with the occurrence DCI-related infarction (p=0.025, r=0.225), it showed no correlation with HbA1c or BG (p=0.699, r=0.043 and p=0.706, r=0.042, respectively). HbA1c failed to show a significant correlation with patient outcome (mRS) after 6 months (p=0.790, r=0.029). In contrast, initial BG correlated with patient outcome after 6 months (p=0.004, r=0.302). Risk for poorer long-term outcome (RR 0.816 95% CI 0.421-1.580, p=0.547) was not increased in patients with elevated HbA1c. However, in patients presenting with hyperglycemia the relative risk for poor outcome (mRS 3-6) increased by 2.19 (p=0.014, 95% CI 1.17-4.11). Despite these correlations, a binary logistic regression model could not show a significant contribution of BG to neurological outcome when compared with known confounders such as WFNS grade and modified Fisher score.
Discussion In this analysis we can demonstrate, that first, elevated HbA1c in patients with aSAH does not correlate with neurological status at admission in contrast to elevated BG. Furthermore, HbA1c is not a relevant predictor for the occurrence of DCI, DCI-related infarction or neurological outcome. Lastly, independent of HbA1c, elevated initial BG serves as a significant predictor of poor outcome after 6 months. Diabetes is associated with an increased frequency and severity of ischemic stroke and hyperglycemia early after stroke onset appears to be an independent contributor to poorer outcome [13]. Comparably, admission hyperglycemia in aSAH is associated with poorer 7
outcome, however, a relevant increase in risk of aneurysm rupture and poorer outcome in diabetic patients could not be demonstrated [1, 3]. The proposed pathophysiological explanation for hyperglycemia attributes this phenomenon to the direct brain injury induced by the hemorrhage through various direct and indirect pathways [2]. Our data illustrates a dissociation of HbA1c and BG after aSAH with a disproportionally higher number of hyperglycemic patients compared to a lower number of patients with pathological HbA1c. Consequently, initial BG appears to correlate more with the severity of the hemorrhage reflected by poorer neurological condition and thus, a possible explanation could be that hyperglycemia may be induced by the acute injury to the brain independent of the pre-ictal degree of glycemia. Though the definition of hyperglycemia remains inconsistent between various studies, elevated BG impacts neurological outcome [1]. In a microdialysis evaluation of cerebral glucose metabolism a threshold for serum BG of 7.5 mmol/l could best distinguish between good and poor outcome and BG > 7.8 mmol/l independently predicted poor outcome [7]. Additionally, occurrence of stress-induced hyperglycemia (a ratio of admission glucose levels and an estimation of preadmission glucose levels) appears to predict necessity for placement of external ventricular drainage [14]. Neither admission blood glucose nor history of diabetes mellitus did increase the risk for symptomatic vasospasm in a retrospective analysis of 352 patients and rates of DCI were comparable between patients with stress-induced hyperglycemia and those without in another study [14, 15]. Concordantly, cerebral perfusion as monitored by early PCT and incidence of DCI were not directly affected by initial hyperglycemia or elevated HbA1c in our patients. Initial hyperglycemia may thus be attributed to metabolic changes caused by early brain injury. However, this relates only to the immediate situation after the hemorrhage, as hyperglycemia in 8
the days following aSAH may increase risk of DCI and aggravates cerebral vasospasm in animal experiment [7, 16, 17]. Limitations This short report holds several limitations. The patient cohort is small and from a single center hindering generalizability. A lack of blinding towards the primary and secondary endpoints exists and a high number of potentially eligible patients had incomplete data sets and were excluded. This leads to a distortion of the patient cohort and contributes to a potentially relevant type-2 error. Moribund patients were excluded from the analysis because in these cases diagnostic and/or therapeutic procedures were often withheld due the very unfavorable prognosis leading to inconsistent data. However, poor grade patients offer interesting insights in the pathophysiology of early brain injury and excluding these patients constitutes a limitation of this analysis. Not all contributing variables known to influence outcome after aSAH were included in our analysis. Especially, modifiable variables are susceptible for therapeutic interventions and exclusion of these variables (e. g. initial blood pressure and use of anticonvulsants) may distort our analysis where neurological outcome was chosen as primary end-point. However, in multivariable logistic and Cox proportional hazards regression initial neurological grade, subarachnoid blood volume and patient age are still among the most relevant contributors and these were included in our analysis [18]. A non-significant trend for higher patient age and poorer neurological grade at admission in patients with elevated HbA1c possibly indicates higher relevance of HbA1c, but this might be 9
obscured by the small cohort size. Additionally, a relevant disproportion in the treatment modality between low and high HbA1c patient groups may have affected end-points.
Conclusion A hyperglycemic metabolism preceding aneurysm rupture, as reflected in an elevated HbA1c, does not contribute to poorer neurological outcome. Hyperglycemia after aSAH correlates with initial neurological status and patient outcome and is potentially attributable to the metabolic changes induced by the brain injury after the hemorrhage.
Acknowledgements None
All listed authors contributed in conceiving the study, analyzing data and preparing and drafting the manuscript. The authors state no conflict of interest and have nothing to disclose. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
10
References 1.
Kruyt, N.D., G.J. Biessels, R.J. de Haan, M. Vermeulen, G.J. Rinkel, B. Coert, et al., Hyperglycemia and clinical outcome in aneurysmal subarachnoid hemorrhage: a metaanalysis. Stroke 2009;40(6):e424-30.
2.
Kruyt, N.D., G.J. Biessels, J.H. DeVries, M.J. Luitse, M. Vermeulen, G.J. Rinkel, et al., Hyperglycemia in aneurysmal subarachnoid hemorrhage: a potentially modifiable risk factor for poor outcome. J Cereb Blood Flow Metab 2010;30(9):1577-87.
3.
Lindgren, A.E., M.I. Kurki, A. Riihinen, T. Koivisto, A. Ronkainen, J. Rinne, et al., Type 2 diabetes and risk of rupture of saccular intracranial aneurysm in eastern Finland. Diabetes Care 2013;36(7):2020-6.
4.
Connolly, E.S., Jr., A.A. Rabinstein, J.R. Carhuapoma, C.P. Derdeyn, J. Dion, R.T. Higashida, et al., Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/american Stroke Association. Stroke 2012;43(6):1711-37.
5.
Teasdale, G.M., C.G. Drake, W. Hunt, N. Kassell, K. Sano, B. Pertuiset, et al., A universal subarachnoid hemorrhage scale: report of a committee of the World Federation of Neurosurgical Societies. J Neurol Neurosurg Psychiatry 1988;51(11):1457.
6.
Frontera, J.A., J. Claassen, J.M. Schmidt, K.E. Wartenberg, R. Temes, E.S. Connolly, Jr., et al., Prediction of symptomatic vasospasm after subarachnoid hemorrhage: the modified fisher scale. Neurosurgery 2006;59(1):21-7; discussion 21-7.
11
7.
Schlenk, F., P. Vajkoczy, A. Sarrafzadeh, Inpatient hyperglycemia following aneurysmal subarachnoid hemorrhage: relation to cerebral metabolism and outcome. Neurocrit Care 2009;11(1):56-63.
8.
Nathan, D.M., H. Turgeon, S. Regan, Relationship between glycated haemoglobin levels and mean glucose levels over time. Diabetologia 2007;50(11):2239-44.
9.
Consensus Statement on the Worldwide Standardization of the Hemoglobin A1C Measurement: The American Diabetes Association, European Association for the Study of Diabetes, International Federation of Clinical Chemistry and Laboratory Medicine, and the International Diabetes Federation. Diabetes Care 2007;30(9):2399-2400.
10.
2. Classification and Diagnosis of Diabetes. Diabetes Care 2016;39 Suppl 1:S13-22.
11.
Vergouwen, M.D., M. Vermeulen, J. van Gijn, G.J. Rinkel, E.F. Wijdicks, J.P. Muizelaar, et al., Definition of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage as an outcome event in clinical trials and observational studies: proposal of a multidisciplinary research group. Stroke 2010;41(10):2391-5.
12.
Etminan, N., K. Beseoglu, H.J. Heiroth, B. Turowski, H.J. Steiger, D. Hanggi, Early perfusion computerized tomography imaging as a radiographic surrogate for delayed cerebral ischemia and functional outcome after subarachnoid hemorrhage. Stroke 2013;44(5):1260-6.
13.
Baird, T.A., M.W. Parsons, P.A. Barber, K.S. Butcher, P.M. Desmond, B.M. Tress, et al., The influence of diabetes mellitus and hyperglycaemia on stroke incidence and outcome. J Clin Neurosci 2002;9(6):618-26. 12
14.
Ray, B., A. Ludwig, L.K. Yearout, D.M. Thompson, B.N. Bohnstedt, Stress-Induced Hyperglycemia After Spontaneous Subarachnoid Hemorrhage and Its Role in Predicting Cerebrospinal Fluid Diversion. World Neurosurg 2017;100:208-215.
15.
Badjatia, N., M.A. Topcuoglu, F.S. Buonanno, E.E. Smith, R.G. Nogueira, G.A. Rordorf, et al., Relationship between hyperglycemia and symptomatic vasospasm after subarachnoid hemorrhage. Crit Care Med 2005;33(7):1603-9; quiz 1623.
16.
Naidech, A.M., K. Levasseur, S. Liebling, R.K. Garg, M. Shapiro, M.L. Ault, et al., Moderate Hypoglycemia is associated with vasospasm, cerebral infarction, and 3-month disability after subarachnoid hemorrhage. Neurocrit Care 2010;12(2):181-7.
17.
Huang, Y.H., C.L. Chung, H.P. Tsai, S.C. Wu, C.Z. Chang, C.Y. Chai, et al., Hyperglycemia Aggravates Cerebral Vasospasm after Subarachnoid Hemorrhage in a Rat Model. Neurosurgery 2017;80(5):809-815.
18.
Rosengart, A.J., K.E. Schultheiss, J. Tolentino, R.L. Macdonald, Prognostic factors for outcome in patients with aneurysmal subarachnoid hemorrhage. Stroke 2007;38(8):231521.
13
Figure legend Figure 1. Flow chart of the patient inclusion and exclusion process
14
Table legends Table 1. Epidemiological data of the included 87 patients stratified into normal and elevated HbA1c levels with test results for group differences. BG denotes blood glucose, WFNS World Federation of Neurological Surgeons Grade, DCI delayed cerebral infarction. Percentages are rounded. Table 2. Epidemiological data of the included 87 patients stratified into normal and elevated blood glucose levels with test results for group differences. BG denotes blood glucose, WFNS World Federation of Neurological Surgeons Grade, DCI delayed cerebral infarction. Percentages are rounded. Table 3. Mean values for HbA1c and blood glucose (BG) levels with standard deviation in brackets and range for all patients and grouped for patients with normal versus elevated HbA1c levels. Group difference was tested using the Student’s t-test corrected for unequal variance.
15
Table 1. Epidemiological data of the included 87 patients stratified into normal and elevated HbA1c levels with test results for group differences. BG denotes blood glucose, WFNS World Federation of Neurological Surgeons Grade, DCI delayed cerebral infarction. Percentages are rounded.
N (%)
All
HbA1c <40mmol/mol
HbA1c >39mmol/mol
87 (100)
61 (70)
26 (30)
56.9 (10.9)
55.5 (10.4)
60.9 (10.9)
0.066 0.630
Age
Years (SD)
Sex
F (%)
57 (66)
41 (67)
16 (62)
M (%)
30 (34)
20 (33)
10 (38)
<7.9mmol/L (%)
47 (54)
33 (54)
14 (54)
>7.8mmol/L (%)
40 (46)
28 (46)
12 (46)
1-3 (%)
48 (55)
37 (61)
11 (42)
4-5 (%)
39 (45)
24 (39)
15 (58)
0 (%)
3 (3)
2 (3)
0
1 (%)
18 (21)
13 (21)
5 (19)
2 (%)
6 (7)
6 (10)
0
3 (%)
45 (51)
30 (49)
15 (58)
4 (%)
16 (18)
10 (16)
6 (23)
Anterior circulation (%)
63 (73)
41 (73)
22 (85)
Posterior circulation (%)
19 (22)
15 (27)
4 (15)
BG
WFNS
Fisher
Aneurysm
Other (%)
5 (5)
16
p
0.983
0.115
0.206
0.461
Treatment
Surgery (%)
45 (52)
36 (61)
9 (35)
Endovascular (%)
40 (46)
23 (39)
17 (65)
None (%) DCI
Infarction
Outcome
0.050
2 (2)
No (%)
51 (59)
35 (57)
16 (62)
Yes (%)
36 (41)
26 (43)
10 (38)
No (%)
73 (84)
51 (84)
22 (85)
Yes (%)
14 (16)
10 (16)
4 (15)
Good (mRS 0-2)
56 (64)
38 (62)
18 (69)
Poor (mRS 3-6)
31 (36)
23 (38)
8 (31)
17
0.718
0.907
0.536
Table 2. Epidemiological data of the included 87 patients stratified into normal and elevated blood glucose levels with test results for group differences. BG denotes blood glucose, WFNS World Federation of Neurological Surgeons Grade, DCI delayed cerebral infarction. Percentages are rounded.
N (%)
All
BG < 7.9 mmol/l
BG > 7.8 mmol/l
87 (100)
47
40
56.9 (10.9)
55.3 (11.2)
59.3 (10.0)
0.034 0.009
Age
Years (SD)
Sex
F (%)
57 (66)
25(44)
32(56)
M (%)
30 (34)
22(73)
8(27)
<40mmol/mol (%)
47 (54)
33(54)
28(46)
>39mmol/mol (%)
40 (46)
14(54)
12(46)
1-3 (%)
48 (55)
32(67)
16(33)
4-5 (%)
39 (45)
15(39)
24(61)
0 (%)
3 (3)
2(100)
0
1 (%)
18 (21)
14(78)
4(22)
2 (%)
6 (7)
3(50)
3(50)
3 (%)
45 (51)
24(53)
21(47)
4 (%)
16 (18)
4(25)
12(75)
Anterior circulation (%)
63 (73)
39(61)
25(39)
Posterior circulation (%)
19 (22)
4(24)
13(76)
HbA1c
WFNS
Fisher
Aneurysm
Other (%)
5 (5)
18
p
1.000
0.010
0.002
0.012
Treatment
Surgery (%)
45 (52)
26(58)
19(42)
Endovascular (%)
40 (46)
19(48)
21(52)
2 (2)
2(100)
0
No (%)
51 (59)
26(51)
25(49)
Yes (%)
36 (41)
21(58)
15(42)
No (%)
73 (84)
41(56)
32(44)
Yes (%)
14 (16)
6(43)
8(57)
Good (mRS 0-2)
56 (64)
36(64)
20(36)
Poor (mRS 3-6)
31 (36)
11(36)
20(64)
None (%) DCI
Infarction
Outcome
0.609
0.521
0.394
0.014
Table 3. Mean values for HbA1c and blood glucose (BG) levels with standard deviation in brackets and range for all patients and grouped for patients with normal versus elevated HbA1c levels. Group difference was tested using the Student’s t-test corrected for unequal variance.
All patients
HbA1c <40mmol/mol
HbA1c >39mmol/mol
87
61
26
HbA1c [mmol/mol]
38.6(±9.9) 25-107
34.7(±2.9) 25-39
47.7(±14.0) 40-107
<0.01
BG [mmol/l]
8.5(±2.5) 4.7-16.7
8.2(±2.1) 4.7-15.4
9.3(±3.1) 5.8-16.7
0.103
N=
19
P=
20