Cardiac autonomic neuropathy (CAN) is one of the most

Transcription

1 14 JAPI september 2012 VOL. 60 Original Article Carotid Intima-Media Thickness in Type 2 Diabetes Mellitus Patients with Cardiac Autonomic Neuropathy Pradip Kumar Sinha *, Gouranga Santra **, Dibyendu De ***, Avishek Saha ***, Kaushik Biswas ***, Prithwiraj Bhattacharya ***, Pramit Ghosh **** Abstract Introduction : Cardiac autonomic neuropathy (CAN) is an important complication of type 2 diabetes mellitus (T 2 ). Accelerated atherosclerosis is also common in T 2. Carotid intima media thickness (CIMT) is a surrogate marker of atherosclerosis. We conducted a study to assess the CIMT in T 2 patients with CAN. Methods : In 84 T 2 patients cardiac autonomic function was assessed by five clinical tests including: 1) heart rate variation during deep breathing, 2) hear rate response to standing, 3) Valsalva ratio, 4) postural fall in systolic blood pressure (BP) three minutes after standing, and 5) resting heart rate. CAN was defined as two or more positive tests out of five for cardiac autonomic function. CIMT was measured by two dimensional (2D) ultrasound. We also examined for presence of any atherosclerotic plaque over intima of carotid artery as well as within the carotid bulb. Results : Thirty six (42.85 percent) out of 84 patients were detected to have CAN. CAN was significantly associated with duration of disease after its detection (P=0.0253), high LDL cholesterol (P=0.0418), low HDL cholesterol (P=0.0001), fasting blood sugar (FBS) level (P=0.0012) and CIMT (P=0.0001) equal to or more than 69 mm. Conclusion : Increased CIMT equal to or more than 69 mm is associated with high occurrence of CAN in diabetic population. Duration of diabetes, abnormal lipid tests and FBS level significantly influence the development of CAN. Introduction Cardiac autonomic neuropathy (CAN) is one of the most neglected chronic complications of diabetes mellitus (). Varying incidence of CAN in have been reported in different studies. Ziegler et al using predefined heart rate variability (HRV) tests and analysis of the R-R intervals found that 25.3% of patients with type 1 diabetes mellitus (T 1 ) and 34.3% of patients with type 2 diabetes mellitus (T 2 ) have abnormal findings suggestive of CAN. 1 Regional sympathetic imbalance, altered vascular responsiveness and QT interval prolongation predispose the patients with CAN to malignant arrhythmias and sudden cardiac death. Silent ischemia is more common in diabetic patients with CAN than in those without CAN (38% vs. 5%). 2 Five year mortality rate is five times higher in diabetic patients having CAN than those without CAN. 3,4 Progressive increase in risk of mortality is seen with increased number of abnormal CAN function tests. 5 So early detection of CAN is of prime importance for risk stratification and intervention for preventing the serious consequences especially in diabetics. As accelerated atherosclerosis and neuropathy (including CAN) in have similar pathogenesis (like activation of polyol pathway, increased glycation end products and increased protein kinase C activity), their correlation is a theoretical possibility. Carotid bulb involvement by atherosclerosis can also explains some features of CAN due to baroreceptor functional abnormality. 6 CAN may itself lead to accelerated atherosclerosis due to increased heart rate and reduced HRV. With this background knowledge, it can be hypothesized that there may be * Associate Professor, ** Assistant Professor, *** Postgraduate Trainee, Dept. of Medicine, **** Assistant Professor, Dept. of Community Medicine, Medical College, 88 College Street, Kolkata Received: ; Revised: ; Accepted: a correlation between CAN and carotid intima-media thickness (CIMT), a surrogate marker of atherosclerosis. A small number of previous studies also showed that CAN in diabetic patients have positive correlation with atherosclerosis and CIMT. 6-8 However studies regarding CIMT with special reference to CAN are still inadequate in diabetic patients. Association of CAN in T 2 patients with the level of thickness of carotid intima-media is not yet well-established. We conducted this cross-sectional study to correlate the level of thickness of carotid intima-media in T 2 patients with development of CAN as well as to identify other factors influencing the development of CAN. Materials and Methods We studied 84 T 2 patients admitted in indoor medical wards over one and half year s duration from July 2008 to December T 2 patients with age <65 years were included in this study. Patients with ongoing angina, congestive cardiac failure, renal disease, thyroid disease, hepatic disease, ketosis, and stroke were excluded from the study. Written consent from all patients was taken before participating in this study and prior approval was obtained from the hospital ethical committee. A detailed history was taken regarding duration and treatment of T 2, hypertension, smoking, and alcohol intake and physical activity. Patients were regarded as physically active who used to do regular aerobic exercise for more than 20 minutes for at least three days per week or walk for 30 minutes regularly or was on a profession that required high level of physical activity, like manual labourer, sports person etc. Clinical examination included measurement of blood pressure (BP), height, weight, waist-hip ratio, fundoscopy and five clinical tests to detect CAN as described later. Blood samples were drawn after 12 hours of fast and fasting blood sugar (FBS), serum urea, creatinine, uric acid and lipid profile were assayed. Blood sugar was estimated

2 JAPI september 2012 VOL using glucoseoxidase-peroxidase (GOD POD) method (Sigma Diagnostics Pvt. Ltd, Vadodara, Gujarat, India). Serum urea was estimated by Urease Berthelot method (Caltek Diagnostics Pvt. Ltd, Kolkata, India). For serum creatinine, alkaline picrate Jaffe s reaction method was used. Serum lipid levels were measured using Hitachi 912 analyzer (Roche Diagnostics, Germany). Doppler study of bilateral carotid arteries was performed in all patients. Imaging was conducted using a high-resolution 7.5 MHz phased-array transducer (Philips IU-22 ultrasound scanner). Imaging of both common carotid arteries upto their bifurcation, the carotid bulb, as well as proximal 10 mm of internal carotid artery of both sides was done by manipulating the transducer in such a way that best images of far wall of the arteries were obtained. Intimal plaques were searched. Sites with intimal plaques were avoided during measurement of CIMT. Mean values of CIMT of three sites of a particular side were taken for calculation of CIMT of that side. Mean values of CIMT of both sides were compared between the two groups i.e. diabetics with CAN and diabetics without CAN. All the recruited patients were subjected to five clinical tests for detecting CAN. Criteria were predefined for each test to be considered as positive for CAN as described below. CAN was considered to be present if two or more criteria were positive. Five clinical tests for detecting CAN: 1. Heart rate variation during deep breathing (positive for CAN if <10/min). 2. Heart rate response to standing: RR interval was measured at beats 15 and 30 after standing during continuous ECG monitoring (positive for CAN if 30:15 ratio <1.03). 3. Valsalva ratio: Patient exhales into mouth-piece of a manometer to 40 mmhg for 15 seconds while on continuous ECG monitoring. Normal subjects develop tachycardia and peripheral vasoconstriction during strain and overshoot bradycardia and BP rise on release. Ratio between longest R-R and shortest R-R < 1.2 was considered positive for CAN (normal >1.2). 4. Postural fall in systolic BP three minutes after standing (positive for CAN if > 30 mm Hg). Table 1 : Distribution of patient parameters Patient Parameters (n=49) (n=35) Total (n=84) p value Mean/freq SD/ percentage Mean/freq SD/ percentage Mean/freq SD/ percentage Age Duration of diabetes FBS PPBS SBP DBP CIMT Total Cholesterol Triglyceride LDL HDL VLDL Smoking % % < % Alcohol % % % CAN positive % % % Patients on Insulin % 21 60% % Patients on Oral hypoglycemic agents (OHA) % 14 40% % 5. Resting heart rate (positive for CAN if >100 bpm). In the early 1970s, Ewing et al proposed five simple noninvasive cardiovascular reflex tests (Valsalva maneuver, heart rate response to deep breathing, heart rate response to standing, blood pressure response to standing, and blood pressure response to sustained handgrip). 9 These tests are still the basic reflex tests of CAN. We included resting heart rate in addition to the initial four criteria, because of its high predictability of presence of CAN. 7 Two groups of patients i.e. diabetics with CAN and diabetics without CAN were compared in their demographic profiles. CIMT was compared between two groups. We also compared the CIMT between male and female patients in total or in subgroups having CAN or no CAN. Statistical method: Epi Info software Version was used for statistical analysis {developed by Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia (USA)}. Datas were expressed in percentages, mean values and standard deviation. In univariate analyses datas were compared using Chi-square test and Fischer s exact test for categorical variables, and student s t test for continuous variables. Multiple logistic regression analysis was performed for assessment of independent contributors of CAN development. P value <0.05 was considered statistically significant. Plotting in receiver operating characteristic (ROC) curve was done to show the most accurate value of CIMT with acceptable sensitivity and specificity rate for detecting development of CAN. Result and Analysis Out of total 84 T 2 patients who were included in the study, 49 were male and 35 were female and all of them fulfilled the inclusion criteria. Their clinical and biochemical characteristics are mentioned in Table 1. Thirty six patients (42.85%) fulfilled our predefined criteria of having CAN. patients (44.9%) were showing slightly higher prevalence of CAN in comparison to females (40%) though the difference was not statistically significant. All other parameters, except history of addiction (smoking and alcohol), were comparable between male and female patients, including duration of diabetes, FBS, PPBS, lipid

3 16 JAPI september 2012 VOL. 60 Table 2 : Clinical and biochemical characteristics of study population: univariate analysis Patient characteristics CAN (n=36) (Mean ± SD)/ % Without CAN (n=48) (Mean ± SD) / % P Value Sex (male/female) 61.11/ / Age (years) ± ± BMI (kg/m 2 ) 27.9 ± ± 3.8 >.05 SBP (mmhg) 160 ± ± DBP (mmhg) 100± ± Total cholesterol ± ± LDL cholesterol ± ± triglyceride ± ± HDL cholesterol ± ± VLDL 47.96± ± Hypertension 94.44% 83.33% Regular smokers 33.33% 22.92% Alcohol intake 16.67% 8.33% Physically active 30.55% 31.25% 1.00 FBS 154.5± ± PPBS ± ± Diabetes duration 8.36± ± (years) CIMT (mm) 0.91± ± Insulin/ OHA 58.30%/41.70% 56.20%/43.80% 1.00 cholesterol level was significantly associated with development of CAN in univariate analysis (Table 2). Differences in other parameters between these two groups including treatment history did not assume any statistical significance (Table 2). CIMT was measured in all patients. Between two groups of patients i.e., patients with CAN (0.91±0.15 mm) and without CAN (0.62±0.08 mm), the difference of CIMT were significant (P =0.0001) (Table 2). s scored higher CIMT than their male counterpart in CAN group (0.99+/-0.18 mm vs /-0.12 mm; p value 0.035), while there is no significant difference among them in non-can group (0.60+/-0.10 mm vs /-0.08 mm; P= 0.347) (Table 4). No significant difference of mean values of CIMT was seen between males and females (0.73+/ mm vs / mm; P=0.612) as a whole (Table 1). CIMT of 0.69 mm in T 2 patients had highly acceptable sensitivity (94%) and specificity (87.5 %) rate for detecting the probability of development of CAN, with a diagnostic accuracy of 90.48% (Figure 1 and Table 5). In patients with CAN the mean age at presentation, systolic blood pressure (SBP), diastolic blood pressure (DBP), total cholesterol, triglyceride and PPBS were higher than patients without CAN, though these differences were not statistically significant (Table 2). Prevalence of smoking habit, addiction to alcohol and sedentary lifestyle were more common in patients with CAN, though statistically were not significant (Table 2). Higher BMI was noted in patients without CAN, though statistically non-significant (p >0.05). Both FBS and PPBS levels were high in majority of our patients during inclusion, despite all of them were receiving antidiabetic therapy. However statistically important association with development of CAN Table 3 : Multiple logistic regression analysis of variables influencing development of CAN Term Odds Ratio 95% C.I. Coefficient S. E. Z-Statistic P-Value Age Duration of diabetes FBS HDL PPBS Total cholesterol Constant * * Table 4 : Comparison of CIMT between male and female patients Patients with CAN (n =36) Patients without CAN (n =48) Total (n =84) n =22 n = 14 n =27 n =21 n =49 n =35 Mean CIMT ± SD P value 0.86± ± ± ± ± ± profile and treatment history (Table 1). Clinical and biochemical parameters of two groups of patients (with CAN and without CAN) are compared in Table 2. We found significant difference between two sub-groups in respect to duration of diabetes after detection and levels of HDL and FBS in multiple logistic regression analysis (Table 3). High LDL was noted only with mean FBS level. Though more patients with CAN were receiving insulin (58.3 %) than those without CAN (56.2 %), this difference was not statistically significant (Table 2). Out of 36 patients in CAN group seven patients had intimal plaque in carotid artery, whereas in non-can group (n=48) only one patient had intimal plaque in carotid artery (p<0.05). Intimal plaque was noted in carotid bulb in two patients with CAN, but no patient in non-can group had intimal plaque in carotid bulb. Discussion CAN was detected in 42.85% of our diabetic patients. Other studies revealed varying incidence of CAN in diabetic populations. Pappachan et al studied one hundred T 1 and T 2 patients and prevalence of CAN was 60% among them. 10 Ziegler et al reported 34.3% of their T 2 patients had CAN at presentation, whereas Nishimura et al reported 21.4% having CAN at presentation. 1,11 Factors which actually play in the development of autonomic neuropathy are still poorly defined. Impact of metabolic changes on neural circulation causing reduced blood flow and hypoxia could be the important factors. 12 Early atherosclerotic involvement of vasanervosum results in nerve injury. Atherosclerotic plaque

4 JAPI september 2012 VOL Sensitivity ROC Curve Specificity Fig. 1 : ROC curve plotting sensitivity and 1-specificity at different CIMT values. Sensitivity of 100% is plotted as 1; specificity of 100% is considered as 1 in this curve. in carotid bulb may also result in disturbances of baro-receptor function. CIMT was correlated to atherosclerosis in various studies and is used as a predictor of future cardiac and cerebro-vascular events particularly in patients of T Progression of CIMT was reported to be 25% greater in diabetic subjects in comparison to non-diabetics. 14 Few studies revealed that CAN in diabetic patients have positive correlation with atherosclerosis. Meyer et al in a study with 45 T 2 patients showed significant association of CAN with arterial dysfunction and hyper-insulinaemia. Markers of early vascular disease like systemic arterial compliance, arterial stiffness and CIMT correlated with total autonomic score. 8 In a study of Gottsatera et al, CAN was associated with features of carotid atherosclerosis detected by ultrasound. They pointed out that carotid atherosclerosis, especially in the carotid bulb, might affect baroreceptor function with progression of T 2. 6 In another study Gottsater et al showed that reduced common carotid artery (CCA) diameter and atherosclerotic intimamedia thickness (IMT) both correlated with HRV at baseline, which itself indicate presence of CAN. 7 A cross-sectional study by Yokoyama et al documented that markers for vascular wall properties such as pulse-wave velocity (PWV), IMT and pulse pressure are significantly altered in diabetic neuropathy patients. 15 In the present study increased CIMT were noted in both CAN and non-can groups. Normal values of age specific thickness level of CIMT is yet to be standardized in South East Asian population. Definitely CIMT will vary depending on different factors like age, sex, ethnicity, associated diseases like, dyslipidemia, whether patients were taking lipid lowering drugs, oral hypoglycemic agents or insulin etc. Different studies reported from India showing different values of CIMT and its Table 5 : Sensitivity and specificity of CIMT value for diagnosing CAN CIMT Sensitivity Specificity PPV NPV Diagnostic Accuracy PPV- positive predictive value; NPV- negative predictive value association with CAD The CUPS study (2000) evaluated CCA-IMT in South Indian subjects and observed that the mean CIMT of the diabetic subjects (0.95 ± 0.31mm) was significantly higher than those of the non-diabetic subjects (0.74 ± 0.14 mm) (p<0.05). 17 Agarwal et al in their study in T2 patients showed the mean IMT in the study group was 0.67± 0.14 mm (males 0.68 ± 0.14 mm and females 0.66 ± 0.15 mm). Mean CIMT in the CAD group was 0.73 ± 0.12 mm as compared to 0.65 ± 0.15 mm in non-cad group (p <.05). 18 Overall mean values of CIMT of all selected diabetic patients in our study was 0.74+/-0.19 mm, which is lower than the diabetic subgroup in CUPS study (2000). 17 In the present study female patients with CAN had statistically high CIMT value than their male counterparts (P <0.035), an unique finding which may reflect accelerated progression of atherosclerosis in female diabetics from Indian subcontinent; a finding which require larger study for confirmation. No significant difference of CIMT was observed between male and females in non-can group. A statistically significant difference was noted between mean values of CIMT in CAN group (0.91+/ mm) and in non-can group (0.62+/-0.08 mm). CIMT of 0.69 mm in T 2 patients had high sensitivity (94%) and specificity (87.5 %) for being associated with CAN. Increased incidence of intimal plaque in carotid artery and carotid bulb in addition to increased CIMT is also associated with CAN. In patients with T 1, duration of disease had been reported to play an important role in relation to development of CAN, 19 but in T 2 the influence of disease duration seems to play a less important role (as actual time of onset of T 2 remains unknown in a majority of patients). 20 Khoharo et al found in their T 2 patients that mean values of known duration of diabetes to develop CAN was 14 ± 3.5 years. 21 In the present study mean duration of disease after diagnosis in CAN group was 8.36 ± 3.1 years and in non-can group was 6.45 ± 1.39 years. A significant association between duration of T 2 after diagnosis and alteration of different clinical tests for diagnosing CAN was noted both in multivariate analysis (P= ) and on univariate analysis (P= ). Nishnianidze et al got high total cholesterol, low HDL and high triglyceride levels in their study population, but they failed to show any significant correlation between CAN severity and lipid profile indices. 22 In present study on univariate analysis high LDL cholesterol (P=0.0418) and low HDL cholesterol (P=0.0001) were found to be significantly associated with development of CAN in T 2 patients. On multiple logistic regression analysis HDL level (p = ) demonstrated a significant negative coefficient ( ) on development of CAN. Also in univariate analysis low HDL level was significantly associated with CAN (P = ). So duration of the disease, high LDL and low HDL cholesterol are found to be individual risk factor for development of CAN in T 2 patients. Ko et al showed that development of CAN was strongly

5 18 JAPI september 2012 VOL. 60 associated with the mean A 1 C level during the follow-up period of their patients (mean A 1 C >9.0% vs. 7.0%, OR 2.984). 23 In Steno-2 trial targeted intensive intervention involving blood sugar and multiple cardiovascular risk factors, reduced the prevalence of CAN among patients with T 2 and microalbuminuria. 24 On the other hand, VA Cooperative Study failed to demonstrate any difference in prevalence of CAN in T 2 patients after two years of tight glycemic control compared with those without tight control. 25 In the present study we got a significant association of FBS level (>120 mg/dl) with development of CAN (P=0.0012), indicating FBS level is a significant risk factor on regression analysis. Conclusion In the present study on multivariate analysis duration of diabetes after diagnosis (P=0.0253), high FBS level (P=0.0012), low HDL cholesterol (P=0.0001), and on univariate analysis high LDL cholesterol (P=0.0418) were all associated with high risk of development of CAN in T 2 patients. Increased CIMT in carotid artery and carotid bulb had a direct correlation with development of CAN in diabetic patients. In this study CIMT 0.69 mm were seen to be significantly associated with development of abnormalities of various CAN function tests in T 2 patients. Studies are needed to establish the role of therapeutic intervention to reduce CIMT level and its effect on CAN. References 1. Ziegler D, Dannehl K, Muhlen H, Spuler M, Gries FA. Prevalence of cardiovascular autonomic dysfunction assessed by spectral analysis, vector analysis, and standard tests of heart rate variation and blood pressure responses at various stages of diabetic neuropathy. Diabet Med 1992;9: Johnston SC, Easton JD. Are Patients with acutely recovered cerebral ischemia more unstable? Stroke 2003;4: Vinik AI, Maser RE, Mitchell BD, Freeman R. Diabetic autonomic neuropathy. Diabetes Care 2003;26: Ziegler D. Cardiovascular autonomic neuropathy: clinical manifestations and measurement. Diabetes Reviews 1999;7: Maser RE, Maser RE, Vinik AI, Freeman R. The association between cardiovascular autonomic neuropathy and mortality in individuals with diabetes: a meta-analysis. Diabetes Care 2003;26: Gottsatera A, Szelagb B, Berglundb G, et al. Changing associations between progressive cardiovascular autonomic neuropathy and carotid atherosclerosis with increasing duration of Type 2 diabetes mellitus. J Diab Complications 2005;19: Gottsater A, Ahlgren AR, Taimour S, Sundkvist G. Decreased heart rate variability may predict the progression of carotid atherosclerosis in type 2 diabetes. Clin Auton Res 2006;16: Meyer C, Milat F, McGrath BP, et al. Vascular dysfunction and autonomic neuropathy in Type 2 diabetes. Diabet Med 2004;21: Ewing DJ, Martyn CN, Young RJ, Clarke BF. The value of cardiovascular autonomic function tests: 10 years experience in diabetes. Diabetes Care 1985;8: Pappachan JM, Sebastian J, Bino BC, et al.cardiac autonomic neuropathy in diabetes mellitus: prevalence, risk factors and utility of corrected QT interval in the ECG for its diagnosis. Postgrad Med J 2008;84: Masato Nishimura, Tetsuya Hashimoto, Hiroyuki Kobayashi, et al. Association between cardiovascular autonomic neuropathy and left ventricular hypertrophy in diabetic haemodialysis patients. Nephrol Dial Transplant 2004;19: Watkins PJ, Edmonds ME. Diabetic autonomic failure. In Mathias CJ, Bannister SR (eds.): Autonomic failure: a textbook of clinical disorders of the autonomic nervous system, 4th ed. New York: Oxford University Press, 1999; Lorenz MW, Markus HS, Bots ML, Rosvall M, Sitzer M. Prediction of clinical cardiovascular events with carotid intima-media thickness: a systematic review and meta-analysis. Circulation 2007; 115: Wagenknecht LE, Zaccaro D, Espeland MA, et al. Diabetes and progression of carotid atherosclerosis: the insulin resistance atherosclerosis study. Arterioscler Thromb Vasc Biol 2003;23: Yokoyama H, Yokota Y, Tada J, Kanno S. Diabetic neuropathy is closely associated with arterial stiffening and thickness in Type 2 diabetes. Diabetic Medicine 2007;24: Jadhav UM, Kadam NN. Carotid intima-media thickness as an independent predictor of coronary artery disease. Ind Heart J 2001;53: Ravikumar R, Deepa R, Shanthirani S, Mohan V. Comparison of carotid intima media thickness, arterial stiffness and brachial artery flow mediated dilatation in diabetic and nondiabetic subjects. The Chennai Urban Population Study (CUPS-9). Am J Cardiol 2001;90: Agarwal AK, Singla Sweta, Singla S, et al. Prevalence of Coronary Risk Factors in Type 2 Diabetics without manifestations of overt coronary heart disease. J Assoc Physicians India 2009;57: Ziegler D. Diabetic cardiovascular autonomic neuropathy: prognosis, diagnosis and treatment. Diabetes Metab Rev 1994;10: Ko SH, Park SA, Cho JH, et al. Progression of cardiovascular autonomic dysfunction in patients with type 2 diabetes-a 7-year follow-up study. Diabetes Care 2008;31: Khoharo HK, Qureshi F. Frequency of cardiac autonomic neuropathy in patients with type 2 diabetes mellitus reporting at a teaching hospital of Sindh. J Coll Physicians Surg Pak 2008;18: Nishnianidze MA, Kurashvili RB, Khelashvili MG, Tsutskiridze LR, Bekaia MG. Cardiac autonomic neuropathy in relation to some components of metabolic syndrome in newly diagnosed type 2 diabetic patients. Presented at European Congress of Endocrinology 2006; Glasgow, UK. Endocrine Abstracts 2007;11: Ko SH, Park SA, Cho JH, et al. Progression of cardiovascular autonomic dysfunction in patients with type 2 diabetes: A 7-year follow-up study. Diabetes Care 2008; 31: Gaede P, Vedel P, Larsen N, et al. Multifactorial intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med 2003;348: Azad N, Emanuele NV, Abraira C, et al. The effects of intensive glycemic control on neuropathy in the VA cooperative study on type II diabetes mellitus. J Diabetes Complications 1999;13: