BioMed CentralBMC Ophthalmology ss Open AcceResearch article Diagonal ear lobe crease in diabetic south Indian population: Is it associated with Diabetic Retinopathy? Sankara Nethralaya Diabetic Retinopathy Epidemiology And Molecular-genetics Study (SN-DREAMS, Report no. 3) Rajiv Raman1, Padmaja Kumari Rani1, Vaitheeswaran Kulothungan2 and Tarun Sharma*1 Address: 1Shri Bhagwan Mahavir Vitreoretinal services, 18, College Road, Sankara mx vir Vitreoretinal services, 18, College Road, Sankara Nethralaya, Chennai-600 006, Tamil Nadu, India and 2Department of Molecular Genetics, 18, College Road, Sankara Nethralaya, Chennai-600 006, Tamil Nadu, India Email: Rajiv Raman - rajivpgraman@gmail.com; Padmaja Kumari Rani - rpk111@gmail.com; Vaitheeswaran Kulothungan - drp@snmail.org; Tarun Sharma* - drtaruns@gmail.com * Corresponding author Abstract Background: To report the prevalence of ear lobe crease (ELC), a sign of coronary heart disease, in subjects (more than 40 years old) with diabetes and find its association with diabetic retinopathy. Methods: Subjects were recruited from the Sankara Nethralaya Diabetic Retinopathy Epidemiology And Molecular-genetics Study (SN-DREAMS), a cross-sectional study between 2003 and 2006; the data were analyzed for the1414 eligible subjects with diabetes. All patients' fundi were photographed using 45° four-field stereoscopic digital photography. The diagnosis of diabetic retinopathy was based on the modified Klein classification. The presence of ELC was evaluated on physical examination. Results: The prevalence of ELC, among the subjects with diabetes, was 59.7%. The ELC group were older, had longer duration of diabetes, had poor glycemic control and had a high socio- economic status compared to the group without ELC and the variables were statistically significant. There was no statistical difference in the prevalence of diabetic retinopathy in two groups. On multivariate analysis for any diabetic retinopathy, the adjusted OR for women was 0.69 (95% CI 0.51-0.93) (p = 0.014); for age >70 years, 0.49 (95% CI 0.26-0.89) (p = 0.024); for increasing duration of diabetes (per year increase), 1.11(95% CI 1.09-1.14) (p < 0.0001); and for poor glycemic control (per unit increase in glycosylated heamoglobin), 1.26 (95% CI 1.19-1.35) (p < 0.0001). For sight-threatening diabetic retinopathy, no variable was significant on multivariable analysis. In predicting any diabetic retinopathy, the presence of ELC had sensitivity of 60.4%, and specificity, 40.5%. The area under the ROC curve was 0.50 (95% CI 0.46-0.54) (p 0.02). Conclusion: The ELC was observed in nearly 60% of the urban south Indian population. However, the present study does not support the use of ELC as a screening tool for both any diabetic retinopathy and sight-threatening retinopathy. Published: 29 September 2009 BMC Ophthalmology 2009, 9:11 doi:10.1186/1471-2415-9-11 Received: 3 February 2009 Accepted: 29 September 2009 This article is available from: http://www.biomedcentral.com/1471-2415/9/11 © 2009 Raman et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Page 1 of 5 (page number not for citation purposes) http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19788727 http://www.biomedcentral.com/1471-2415/9/11 http://creativecommons.org/licenses/by/2.0 http://www.biomedcentral.com/ http://www.biomedcentral.com/info/about/charter/ BMC Ophthalmology 2009, 9:11 http://www.biomedcentral.com/1471-2415/9/11 Background Coronary heart disease (CHD) is a leading cause of mor- tality in persons with type 2 diabetes [1]. In the general population, macrovascular disease is the primary patho- genic mechanism causing CHD; however, in population with diabetes, it is the microvascular disease, in addition to the macrovascular component, that may play a role in the development of CHD. Recent evidences from the ARIC (Atherosclerosis Risk In Communities) study sug- gest that the presence of diabetic retinopathy (microvas- cular disease) poses a two-fold higher risk of CHD (macrovascular disease) and a three-fold higher risk of fatal CHD; therefore, there is a link between macrovascu- lar and microvascular pathogenic mechanisms [2]. The presence of ear lobe crease (ELC) and its association with CHD was first described in 1973 [3]. Blodgett et al found that 75% of CHD cases had ear lobe crease as com- pared to 35% of the controls (age and gender matched) [4]. Hence, there is a link between ELC and macrovascular disease (CHD), and between, microvascular disease and macrovascular disease. But what link we have between ELC and microvascular disease, we do not know. The present study is aimed to find out the prevalence of ELC in south Indian diabetic population and its association with diabetic retinopathy in a population-based study. Methods The study design and research methodology of SN- DREAMS 1 are described in detail elsewhere [5]. The sam- ple was stratified based on the socio-economic scoring: low (score, 0-14), middle (score, 15-28), and high (score, 29-42) [6]. The scoring was calculated on the basis of several param- eters such as ownership/type of residence (rent or own), number of rooms in the house, educational status, salary, occupation, material possessions (cycle, TV, audio, car etc.) and house/land value. Eligible patients, above 40 years, were enumerated using the multistage random sam- pling method. For all those who were not previously diag- nosed to have diabetes diabetic, fasting blood glucose estimation was done twice: first, in the field using capil- lary blood, and second, at the base hospital using labora- tory method (glucose oxidase method) [7]. Patients were considered to be newly diagnosed with diabetes, if the fasting blood glucose level was ≥110 mg/dl on two occa- sions, as described above [8]. The study was approved by the Institutional Review Board, and an informed consent was obtained from all individuals. Demographic data, socio-economic status, physical activ- ity, risk of sleep apnea, dietary habits, and anthropometric measurements were collected. A detailed medical and ocular history and a comprehensive eye examination, including stereo fundus photographs, were taken at the base hospital. Biochemical investigations (Blood sugar, total serum cholesterol, high-density lipoproteins, serum triglycerides, hemoglobin, glycosylated hemoglobin HbA1c) were conducted at the base hospital in fasting state. The presence of a diagonal ear lobe crease was assigned to a person with a crease stretching obliquely from the outer ear canal towards the border of the ear lobe of both the ears; examiners were guided to diagnose the ELC, by com- paring the features with the standard photograph pro- vided to them. Diabetic retinopathy was clinically graded using Klein's classification (Modified Early Treatment Diabetic Retin- opathy Study scales) [9]. The alternative method involves grading all stereoscopic standard fields as a whole, and assigning a level of severity for the eye according to the greatest degree of retinopathy using a modified Airlie House Classification scheme. Retinal photographs were taken after pupillary dilatation (Carl Zeiss Fundus Cam- era, Visucamlite, Jena, Germany); all patients underwent 45° four-field stereoscopic digital photography (posterior pole, nasal, field, superior, and inferior). All photographs were graded by two independent observers in a masked fashion; the grading agreement was high (k = 0.83) [5]. Sight-threatening diabetic retinopathy was defined as the presence of severe non-proliferative diabetic retinopathy, proliferative diabetic retinopathy and clinically significant macular edema [10]. Of the 5,999 subjects enumerated, 5,778 (96.32%) responded for the first fasting blood sugar estimation. Of the 5,778 subjects, 1,816 individuals (1,349 with known history of diabetes and 467 with provisionally diagnosed diabetes) were invited to visit the base hospital. Of the 1,563 (85.60%) who responded, 138 were excluded; in two, the age criteria was not met, and in 136, the second fasting blood sugar was ≤ 110 mg/dl). Provisionally diag- nosed diabetes was defined as new asymptomatic individ- ual with a first fasting blood glucose level ≥ 110 mg/dl. An additional 11 individuals were excluded as their digital fundus photographs were of poor quality, making them ungradable for further analysis. Thus, a total of 1,414 indi- viduals were analyzed for the study. Results Of the 1,414 subjects who were analyzed for diabetic retinopathy, the mean age was 56.3 ± 10 yrs; 750 (53.04%) were men, and 664 were (46.96%), women. The prevalence of ELC in the diabetic population was 844 (59.7%) (95% CI 57.1 - 62.2)); ELC was evident in both the ears, in all subjects. Diabetic retinopathy was seen in 255 (18.03%) (95%CI 16.06-20.13) subjects. Page 2 of 5 (page number not for citation purposes) BMC Ophthalmology 2009, 9:11 http://www.biomedcentral.com/1471-2415/9/11 Table 1 compares the clinical and laboratory data between the ELC group and the group without ELC. The subjects in ELC group (844) were older, had longer duration of dia- betes, had poor glycemic control, and had a high socio- economic status as compared to the group without ELC (570); all variables were statistically significant. No differ- ences were observed between the two groups with regard to gender, smoking and alcohol, diabetic retinopathy, BMI, Blood pressure, serum lipids and microalbuminuria. Table 2 shows the results of logistic regression analysis, keeping diabetic retinopathy as an outcome variable. For any diabetic retinopathy, the univariate analyses identi- fied several associated factors: women OR 0.64 (95% CI 0.49-0.85) (p = 0.002), for 50-59 years, OR was 1.7 (95% CI 1.18-2.44) (p = 0.012), for 60-69 years, OR was 1.7 (95% CI 1.13-2.45) (p = 0.010), increasing duration of diabetes per year OR 1.1 (95% CI 1.09-1.13) (p < 0.0001), and poor glycemic control (per unit increase in glyco- sylated heamoglobin) OR 1.3 (95% CI 1.21-1.36) (p < 0.0001). In multivariate analysis, the adjustments were done for age, gender, duration of diabetes, serum lipids, blood pressure, socio-economic status and glycemic con- trol. From the multivariate analysis, the adjusted OR for women was 0.69 (95% CI 0.51-0.93) (p = 0.014); for age >70 years, 0.49 (95% CI 0.26-0.89) (p = 0.024); for increasing duration of diabetes (per year increase), 1.11(95% CI 1.09-1.14) (p < 0.0001); for presence of ELC, 0.88 (95% CI 0.65-1.19) (p = 0.422) and for poor glycemic control (per unit increase in glycosylated heamo- globin), 1.26 (95% CI 1.19-1.35) (p < 0.0001). For sight-threatening diabetic retinopathy, the univariate analysis revealed the following risk factors: presence of ear lobe crease OR was 2.02 (95% CI 1.02-4.73) (p = 0.04) and the OR for the increasing duration of diabetes (per year) was 1.05 (95% CI 1.00-1.10) (p = 0.045). No varia- ble was significant on the multivariable analysis. The adjusted OR for ELC was 2.00 (95% CI 0.91-4.35) (p = 0.086). Compared to the gold standard — photographic classifi- cation — in diagnosing sight-threatening diabetic retin- opathy, the presence of ELC had a sensitivity of 75%, and specificity of 42.3%. The positive predictive value was only 19.8%. The area under the ROC curve was 0.59 (95% Table 1: Clinical and Laboratory characteristics in ELC group versus non-ELC group ELC Group (Present) n = 844 ELC Group (absent) n = 570 P value Age (years) 57.39+/-10.13 54.75+/-9.65 <0.0001 Gender Male 463 (54.9) 287 (50.4) 0.1076 Female 381 (45.1) 283 (49.6) 0.1076 Smoking 169 (20.0) 108 (18.9) 0.6576 Alcohol 175 (20.73) 135 (23.68) 0.2114 Retinopathy group No DR 690 (81.8) 469 (82.3) 0.8654 Any DR 154 (18.2) 101 (17.71) 0.8657 NST 124 (14.7) 91 (16.0) 0.5538 ST 30 (3.6) 10 (1.8) 0.0680 Duration of DM (months) 79.90+/-76.66 70.86+/-73.91 0.0275 BMI (Kg/m2) 25.48+/-4.16 25.22+/-3.98 0.2410 BP Systolic (mmHg) 139.51+/-21.13 138.57+/-20.27 0.4044 BP Diastolic (mmHg) 81.98+/-11.33 81.93+/-11.44 0.9354 Glycosylated Hb 8.34+/-2.25 7.99+/-2.11 0.0033 Serum cholesterol (mg/dL) 186.77+/-39.77 186.16+/-42.19 0.7826 Serum HDL (mg/dL) 39.35+/-9.66 39.06+/-10.91 0.5994 Ratio HDL/cholestrol 0.200+/-0.059 0.204+/-0.064 0.2272 SES Lower 653 (77.36) 500 (87.71) <0.0001 Upper 191 (22.6) 70 (12.3) <0.0001 Microalbuminuria Normal 681 (80.7) 469 (82.3) 0.4914 Microalbuminuria 139 (16.5) 87 (15.3) 0.6670 Clinical microbuminuria 24 (2.8) 14 (2.5) 0.8612 ELC: Ear lobe crease, DR: Diabetic retinopathy, NST: Non-Sight threatening diabetic retinopathy including mild and moderate non-proliferative diabetic retinopathy, ST: Sight threatening diabetic retinopathy including severe non-proliferative diabetic retinopathy, proliferative diabetic retinopathy and diabetic macular edema, DM: Diabetes mellitus, BMI: Body mass index, BP: Blood pressure, Hb: Heamoglobin, HDL: High density lipoprotein, SES: Socioeconomic status. Page 3 of 5 (page number not for citation purposes) BMC Ophthalmology 2009, 9:11 http://www.biomedcentral.com/1471-2415/9/11 CI 0.50-0.68) (p = 0.047). In predicting any diabetic retin- opathy, the presence of ELC had sensitivity of 60.4%, and specificity, 40.5%. The area under the ROC curve was 0.50 (95% CI 0.46-0.54) (p 0.02). Discussion The diagonal ELC has been suggested as a simple marker of vascular disease in the general population, but in pop- ulation with diabetes (population with increased risk of microangiopathy) only limited data are available [11]. The Fremantle diabetes study reported the prevalence of ELC to be 55% in the western Australian population [11]. Our data suggest that the ELC was present in 59.7% of the diabetic population > 40 years in the urban south Indian population. The present study shows that the subjects in the ELC group were older, had longer duration of diabetes and had poor glycemic control. Similar observations were found in the Fremantle study [11]. Subjects with ELC had a higher socio-economic status as compared to the group without ELC; this could be an indirect measure of the population that is at a greater risk for coronary artery disease. With regards to the association between ELC and any dia- betic retinopathy, we noted that increasing age, poor gly- cemic control and increasing duration of diabetes were significant variables in both univariate and multivariate models. Similar observations were made in other popula- tion-based studies on diabetic retinopathy [12-15]. Regarding the association between ELC and sight-threat- ening diabetic retinopathy, the univariate analysis showed that subjects with ELC had almost twice the risk of developing sight-threatening diabetic retinopathy. The possible explanations for this association include loss of elastin; this might be responsible for ear lobe crease and similar loss of elastin in a retinal blood vessel might account for increased leakage and dilatation. The other speculation is the ischemia; the focal ischemia of the der- mal fat might cause ear lobe crease, and ischemia in retina causes sight- threatening changes. However, these specu- lations need further studies. Further, in multivariate anal- ysis, when the effect of other variables was adjusted, the association between the ELC and the sight-threatening diabetic retinopathy was not significant. Taking into con- sideration the low sensitivity and specificity value along with low positive predictive value, the presence of ELC Table 2: Regression Analysis to study the effect of various risk factors on Diabetic Retinopathy and Sight-threatening Diabetic Retinopathy. Diabetic Retinopathy Sight-threatening Diabetic Retinopathy Univariate Analysis Multivariate analysis Univariate Analysis Multivariate analysis OR 95% CI p OR 95% CI p OR 95% CI p OR 95% CI p Gender Men 1 1 1 1 Women 0.64 0.49-0.85 0.002 0.69 0.51-0.93 0.014 0.86 0.42-1.73 0.667 0.92 0.44-1.91 0.822 SES Lower 1 1 1 1 Upper 0.09 0.63-1.29 0.584 0.72 0.48-1.07 0.104 1.24 0.53-2.92 0.617 1.00 0.39-2.52 0.996 Ear lobe crease Absent 1 1 1 1 Present 1.036 0.79-1.37 0.800 0.88 0.65-1.19 0.422 2.02 1.02-4.73 0.043 2.00 0.91-4.34 0.086 Age groups (years) 40-49 1 1 1 1 50-59 1.698 1.18-2.44 0.012 1.38 0.94-2.04 0.100 2.39 0.76-7.15 0.135 2.13 0.66-6.80 0.204 60-69 1.665 1.13-2.45 0.010 1.08 0.71-1.66 0.710 2.86 0.88-9.24 0.079 2.43 0.73-8.08 0.148 ≥ 70 1.120 0.66-1.88 0.673 0.49 0.26-0.89 0.024 3.22 0.78-13.30 0.106 2.83 0.65-12.32 0.167 Duration of Diabetes (per year increase) 1.110 1.09-1.13 <0.0001 1.11 1.09-1.14 <0.0001 1.050 1.00-1.10 0.046 1.04 0.99-1.09 0.123 HbA1c (per unit increase) 1.280 1.21-1.36 <0.0001 1.26 1.19-1.35 <0.0001 1.060 0.93-1.22 0.383 1.07 0.93-1.25 0.343 OR: Odds Ratio, SES: Socioeconomic status, CI: Confidence interval, HbA1c: Glycosylated heamoglobin Page 4 of 5 (page number not for citation purposes) BMC Ophthalmology 2009, 9:11 http://www.biomedcentral.com/1471-2415/9/11 Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical researc h in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp BioMedcentral cannot be used as a screening tool to predict diabetic retinopathy, including sight-threatening diabetic retinop- athy. The strength of this study was that it used photography and standard grading techniques for diabetic retinopathy. Further, the study was representative of a large population and results could be extrapolated to the whole of urban India. One of the limitations of the study was not studying the non-diabetic population with regards to ELC; it would have been interesting to compare the findings with sub- jects without diabetes. The second limitation was not grading ear lobe crease; recent evidence has pointed out to the relationship between the increasing grades of ELC and the increasing severity of coronary artery disease[16]. Thirdly, the sample in the subgroup of sight-threatening diabetic retinopathy is small; -thus, a lower power in this subgroup analysis. Conclusion ELC is present in nearly 60% of urban south Indian pop- ulation with diabetes, aged above 40 years. The presence of ELC is somewhat related to sight-threatening diabetic retinopathy on univariate analysis. Competing interests The authors declare that they have no competing interests. Authors' contributions RR carried out the clinical evaluation in the study and wrote the manuscript. PKR participated in the design of the study, VK performed the statistical analysis and TS conceived the study, and participated in its design and coordination. All authors read and approved the final manuscript. Acknowledgements RD TATA Trust, Mumbai, India for funding the study. References 1. Ghaffar A, Reddy KS, Singhi M: Burden of non-communicable diseases in South Asia. BMJ 2004, 328:807-10. 2. Klein R, Sharrett AR, Klein BE, Moss SE, Folsom AR, Wong TY, Bran- cati FL, Hubbard LD, Couper D, ARIC Group: The association of atherosclerosis, vascular risk factors, and retinopathy in adults with diabetes: the atherosclerosis risk in communities study. Ophthalmology 2002, 109:1225-34. 3. Frank ST: Aural sign of coronary-artery disease. N Engl J Med 1973, 289:327-8. 4. Blodgett G: The presence of a diagonal ear-lobe crease as an indicator of coronary artery disease, thesis. University of Utah, Salt Lake City; 1983. 5. Agarwal S, Raman R, Paul PG, Rani PK, Uthra S, Gayathree R, McCarty C, Kumaramanickavel G, Sharma T: Sankara Nethralaya-Diabetic Retinopathy Epidemiology and Molecular Genetic Study (SN-DREAMS 1): study design and research methodology. Ophthalmic Epidemiol 2005, 12:143-53. 6. Oakes JM, Rossi PH: The measurements of SES in health research: current practice and steps towards a new approach. Soc Sci Med 2003, 56:769-784. 7. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus: Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2003, 26(suppl):S5-20. 8. Sharma T: How necessary is the second estimation of fasting plasma glucose level by laboratory venous blood to diagnose Type 2 diabetes, particularly in epidemiological studies? Oph- thalmic Epidemiol 2006, 13:281-2. 9. Klein R, Klein BEK, Magli YL, et al.: An alternative method of grading diabetic retinopathy. Ophthalmology 1986, 93:1183-7. 10. Baeza M, Orozco-Beltrán D, Gil-Guillen VF, Pedrera V, Ribera MC, Pertusa S, Merino J: Screening for sight threatening diabetic retinopathy using non-mydriatic retinal camera in a primary care setting: to dilate or not to dilate? Int J Clin Pract 2009, 63(3):433-8. 11. Davis TM, Balme M, Jackson D, Stuccio G, Bruce DG: The diagonal ear lobe crease (Frank's sign) is not associated with coronary artery disease or retinopathy in type 2 diabetes: the Freman- tle Diabetes Study. Aust N Z J Med 2000, 30:573-7. 12. Xie XW, Xu L, Jonas JB, Wang YX: Prevalence of diabetic retin- opathy among subjects with known diabetes in China: the Beijing Eye Study. Eur J Ophthalmol 2009, 19(1):91-9. 13. Raman R, Rani PK, Reddi Rachepalle S, Gnanamoorthy P, Uthra S, Kumaramanickavel G, Sharma T: Prevalence of diabetic retinop- athy in India: Sankara Nethralaya Diabetic Retinopathy Epi- demiology and Molecular Genetics Study report 2. Ophthalmology 2009, 116(2):311-8. 14. Klein R, Knudtson MD, Lee KE, Gangnon R, Klein BE: The Wiscon- sin Epidemiologic Study of Diabetic Retinopathy: XXII the twenty-five-year progression of retinopathy in persons with type 1 diabetes. Ophthalmology 2008, 115(11):1859-68. 15. Lim A, Stewart J, Chui TY, Lin M, Ray K, Lietman T, Porco T, Jung L, Seiff S, Lin S: Prevalence and risk factors of diabetic retinopa- thy in a multi-racial underserved population. Ophthalmic Epide- miol 2008, 15(6):402-9. 16. Higuchi Y, Maeda T, Guan JZ, Oyama J, Sugano M, Makino N: Diago- nal earlobe crease are associated with shorter telomere in male Japanese patients with metabolic syndrome. Circ J 2009, 73(2):274-9. Pre-publication history The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1471-2415/9/11/prepub Page 5 of 5 (page number not for citation purposes) http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=15070638 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=15070638 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=12093643 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=12093643 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=12093643 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=4718047 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=16019696 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=16019696 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=12560010 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=12560010 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=12560010 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=12502614 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=12502614 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=16877287 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=16877287 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=16877287 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=3101021 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=3101021 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19222628 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19222628 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19222628 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=11108067 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=11108067 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=11108067 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19123155 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19123155 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19123155 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19084275 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19084275 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19068374 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19068374 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19068374 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19065433 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19065433 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19060421 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19060421 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Abstract&list_uids=19060421 http://www.biomedcentral.com/1471-2415/9/11/prepub http://www.biomedcentral.com/ http://www.biomedcentral.com/info/publishing_adv.asp http://www.biomedcentral.com/ Abstract Background Methods Results Conclusion Background Methods Results Discussion Conclusion Competing interests Authors' contributions Acknowledgements References Pre-publication history