Research Article | Open Access
Urinary Leukotriene E4 Is Associated with Renal Function but Not with Endothelial Function in Type 2 Diabetes
Leukotrienes are inflammatory and vasoactive mediators implicated in endothelium-dependent relaxations and atherosclerosis. Urinary leukotriene E4 (U-LTE4) is a validated disease marker of asthma and increases also in diabetes and acute coronary syndromes. The aim of the present study was to evaluate the association of U-LTE4 and CRP with endothelial function. Urine samples were obtained from 30 subjects (80% males; median age 65) with type 2 diabetes of at least two years duration and a median glomerular filtration rate (eGFR) of 71 (14–129) mL/min. Reactive hyperemia index (RHI) was used as a measure of microvascular endothelial function, whereas macrovascular endothelial function was determined be means of flow-mediated dilatation of the brachial artery (FMD). Decreased renal function was associated with lower concentrations of U-LTE4. In addition, U-LTE4 was correlated with serum creatinine (; ) and eGFR (; ). A stepwise multiple linear regression analysis identified eGFR as an independent predictor of U-LTE4 concentrations. In conclusion, the present results did not establish an association of U-LTE4 with endothelial dysfunction. However, eGFR was an independent predictor of U-LTE4, but not CRP, in this cohort, suggesting that GFR should be considered in biomarker studies of U-LTE4.
Urinary concentrations of the lipid-derived mediator leukotriene E4 (U-LTE4) may serve as a disease marker in several pathological conditions, such as asthma , obstructive sleep apnea , and acute coronary syndromes . In addition, U-LTE4 has been associated with diabetes , and U-LTE4 also increases gradually with body mass index (BMI) . In the latter contexts, the leukotriene pathway may not only serve as diagnostic markers but could potentially also offer mechanistic insights into the role of inflammation in metabolic disease. For example, targeting either leukotriene receptor signaling or leukotriene synthesis decreases proinflammatory cytokine secretion from visceral fat  and protects against insulin resistance in diet-induced obesity [6, 7].
Importantly, type 2 diabetes is associated with endothelial dysfunction, which plays an important role in the development of vascular complications in these patients . The abnormalities in vascular reactivity associated with type 2 diabetes concern both the micro- and macrocirculation . In the microcirculation, endothelial dysfunction is reflected by reduced extremity skin hyperemia, whereas macrovascular endothelial function can be assessed through brachial artery flow-mediated vasodilatation (FMD).
The leukotriene pathway is activated during the atherosclerosis process [10, 11] and affects vascular function both in vitro and in vivo . Although no previous study has specifically assessed U-LTE4 as a marker of endothelial function, there are several indications for a role of the leukotriene pathway in endothelium-dependent vascular reactivity. For example, leukotrienes induce endothelium-dependent responses in isolated human vessels , through the release of nitric oxide  and prostacyclin . In addition, salivary leukotriene concentrations were recently associated with an increased pulse wave velocity , which may serve as an indirect measure of global endothelial function .
Based on the above observations, the aim of the present study was to assess U-LTE4 in relation to micro- and macrovascular endothelial function in subjects with type 2 diabetes and to compare the results with the established inflammatory marker CRP. The results indicate that in this patient group, renal function is an important confounder for the association of U-LTE4, which should be taken into account when using this biomarker.
2. Materials and Methods
The present cohort represents the baseline examination of a randomised, double-blind, and placebo-controlled study assessing the effect of the endothelin receptor antagonist bosentan treatment on endothelial function. Fourty-six patients with type 2 diabetes of at least two years duration and microalbuminuria were enrolled into the study, and analyzable urine samples were obtained from 30 subjects for measures of U-LTE4. The main study has been described elsewhere . In brief, patients with type 2 diabetes of at least two years duration and microalbuminuria were recruited from the department of Endocrinology, Metabolism and Diabetes at the Karolinska University Hospital (Stockholm, Sweden) from autumn 2007 to spring 2010. Patients were classified as having diabetes mellitus if fasting blood glucose exceeded 7.0 mmol/L (at least on two occasions) or blood glucose concentration was >11.0 mmol/L two hours after an oral glucose loading (75 g). Albuminuria was defined as urine albumin concentration >20 g/L or >30 g/12 h and a ratio of albumin/creatinine >3.0 mg/mmol. Patients were excluded if they had a myocardial infarction or unstable angina within the last three months, decompensated heart failure, changed dose of any vasodilator drug during the preceding six weeks, childbearing potential, impaired hepatic function (defined as twice the normal limit of serum aminotransferases), ongoing treatment with glibenclamide, cyclosporine, or warfarin, or concomitant disease that may have interfered with the possibility for the patients to comply with or complete the study protocol. The study protocol was conducted according to the Declaration of Helsinki and was approved by the local ethics review board. Informed consent was obtained from all patients. The baseline investigations before randomization consisted of assessment of microvascular and macrovascular endothelial function and blood and urine sampling.
2.2. Peripheral Endothelial Function Testing
The investigations were performed following a light breakfast in the morning after 20 min rest with the patient in the supine position. The patients were asked to refrain from caffeine-containing drinks or tobacco consumption and intake of drugs was withheld for the 12 preceeding hours and until after the examinations. Noninvasive determination of digital endothelial function was measured with a pulse amplitude tonometry (PAT) device placed on the tip of each index finger (Endo-PAT2000, Itamar Medical, Caesarea, Israel). The PAT device comprises a pneumatic plethysmograph that applies a uniform pressure to the surface of the distal finger, allowing measurement of pulse volume changes. The inflation pressure of the digital device was electronically set to 10 mm Hg below diastolic blood pressure or 70 mm Hg (whichever was lower). The PAT signal was recorded at baseline and following 5 min arterial occlusion using an inflatable cuff while the contralateral arm served as a control. The blood pressure cuff was inflated 60 mm Hg higher than the systolic pressure or at least 200 mm Hg for 5 minutes. Lack of residual pulsatility was monitored throughout the occlusion period. The postocclusive hyperemia stimulates endothelium-dependent vasodilatation causing an increase in digital pulse amplitude [18, 19]. Pulse amplitude was recorded electronically in both fingers and analyzed by a computerized, automated algorithm (Itamar Medical). The change from the baseline measurement is expressed as the reactive hyperemia index (RHI) which in part reflects vasodilator function of the digital microcirculation . Previous evaluation of this method has demonstrated that RHI is to a large part dependent on NO bioavailability . Endothelium-independent vasodilatation (EIDV) was determined following sublingual administration of nitroglycerine (0.4 mg). The ratio between the maximal pulse amplitude following nitroglycerine administration and baseline pulse amplitude was used for calculating EIDV. The coefficient of variation for Endo-PAT is 12% in our laboratory.
2.3. Flow Mediated Vasodilatation of the Brachial Artery
Macrovascular endothelial function was determined by analysis of flow-mediated dilatation (FMD) of the brachial artery during reactive hyperemia following forearm ischemia described in detail earlier  by means of an 8 MHz linear-array transducer connected to an Acuson Sequoia (Acuson Corporation, Mountain View, CA, USA) in parallel with the determination of digital PAT. Baseline images were saved every third second during one minute and a mean value was calculated from these values. The artery was continuously imaged for three minute during the hyperemia following release of the pressure cuff to determine endothelium-dependent vasodilatation. A mean value was calculated from three recordings at maximum dilatation. EIDV was determined following sublingual administration of nitroglycerine. All images were analyzed using proprietary software (Brachial analyzer, Medical Imaging Applications, Iowa City, IA, USA) by a an experienced technician. The coefficient of variation for FMD is 19% in our laboratory.
2.4. Laboratory Investigations
Fasting plasma glucose, Hb1Ac (Mono S), and blood lipids were assessed with standard methods according to local laboratory routines. High-sensitive C-reactive protein was analyzed using turbidimetry (Beckman Coulter, Fullerton CA, USA). Urine was collected in the morning and frozen in aliquots at −80°C until analysis. Urinary creatinine was determined using a standard colorimetric assay. U-LTE4 was measured using enzyme immune assay kits from Cayman Chemicals (Ann Arbor), and results were expressed as pg U-LTE4 per mg U-creatinine as previously described . Glomerular filtration rate (eGFR) was estimated using the Cockroft-Gault formula.
2.5. Statistical Analysis
Data are expressed as median (range) and frequencies are expressed as percentages. Univariate correlations between either U-LTE4 or hsCRP and clinic parameters were established by Spearman correlation. A multiple stepwise linear regression was performed to evaluate independent predictors of U-LTE4 and hsCRP, respectively. was considered significant. Analyses were performed using SigmaPlot version 12 (Systat Software Inc.).
3. Results and Discussion
The clinical and biochemical characteristics of the 30 study subjects (80% males) of the study cohort are shown in Table 1. Stratifying the subjects according to renal function, measured as eGFR, revealed significant differences for flow mediated vasodilatation, consistent with a decreased macrovascular endothelial function in the group exhibiting severe renal function with eGFR <60 mL/min . Of the inflammatory markers, hsCRP remained unchanged between the groups, whereas U-LTE4 exhibited significantly decreasing concentrations with decreased renal function (Table 1).
Data shown as median (min, max).|
The two inflammatory markers hsCRP and U-LTE4 exhibited a positive correlation (), which however did not reach statistical significance (). The univariate analyses for the respective disease markers with the clinical and biochemical characteristics of the study cohort are shown in Table 2. Lipid measurements did not correlate with either hsCRP or U-LTE4 (Table 2). There were no significant correlations for hsCRP with any of the tested parameters in either the univariate (Table 2) or multivariate (data not shown) analysis. CRP is a well-established biomarker for systemic low grade inflammation and has been associated with arterial stiffness , intima media thickness , subclinical carotid atherosclerosis , and metabolic syndrome . Patients with type 2 diabetes have increased hsCRP which decreases with treatment intensity and correlate positively with a drop in HbA1c .
Inflammation may be a key feature of endothelial dysfunction. Previous studies have reported a significant inverse association between microvascular endothelial function during reactive hyperemia and hsCRP in men with stable coronary disease and macrovascular endothelial function measured by FMD in another study [28, 29]. Nevertheless, the lack of correlations with macrovascular endothelial function in the present study is in line with a previous larger cohort of healthy individuals in which brachial artery FMD had no relationship with hsCRP . Furthermore, the present study did not reveal any associations of U-LTE4 with either macro- or microvascular endothelial function (Table 2).
Several potential confounders should be taken into account when assessing the constituents of the leukotriene pathway as disease markers. For example, an association between U-LTE4 and obesity was initially demonstrated in subjects with obstructive sleep apnea  and was subsequently reported in asthmatics . Likewise, salivary concentrations of leukotriene B4 (S-LTB4) associated with BMI and waist to hip ratio in a case control study of hypertension . In contrast to those studies, the present study did not reveal any significant associations between U-LTE4 and either BMI or waist circumference (Table 2). Likewise, the present cohort lacked an association between U-LTE4 and HbA1c (Table 2), which support the findings of a previous study in which U-LTE4 remained unchanged before and after intense glycemic control in type 2 diabetes . It should be noted that the U-LTE4 concentrations in the present study were higher compared with what was previously measured in healthy controls and in younger diabetics [4, 33], but similar to those reported patient cohorts of asthma , renal failure , or acute coronary syndrome . In addition to dissimilar age and pathologies, also alterations of creatinine clearance and the fact that U-LTE4 was measured using different techniques (ELISA versus liquid chromatographic tandem mass spectrometry) could contribute to higher creatinine-normalized U-LTE4 concentrations in the present study.
Concentrations of U-LTE4 were significantly correlated with measures of renal function in terms of plasma creatinine and eGFR (Table 2). The multivariate analysis including all parameters listed in Table 2 confirmed these findings, showing that eGFR was an independent predictor of U-LTE4 concentrations in these patients (-coefficient 0.325; ).
In a rodent model of glomerulonephritis, urinary leukotriene concentrations were elevated at initial stages of disease, followed by a decline at more advanced phases . In addition, renal function is ameliorated by anti-leukotriene treatment in several animal models of renal failure, including proteinuria in diabetic rats , renal ischemia reperfusion , and drug-induced nephrotoxicity . Previous studies of U-LTE4 in relation to renal function in humans have, however, generated conflicting results. In a study of patients with hepatorenal syndrome, U-LTE4 excretion rate fell in parallel with creatinine clearance, indicative of a GFR-dependent renal excretion of this inflammatory marker . In contrast, in another study of kidney transplant recipients, U-LTE4 was not associated with either serum creatinine or transplant rejection . In the present study of patients with type 2 diabetes of at least two years duration and microalbuminuria, renal function was an independent predictor of U-LTE4 even when taking into account anthropometric and biochemical variables as well as measures of micro- and macrovascular endothelial function.
In summary, the present study did not reveal any significant associations of U-LTE4 with measures of endothelial function and identified renal function as an independent predictor of U-LTE4. Several limitations must, however, be acknowledged. This study represents a limited sample size. In addition, diabetes patients were selected based on existing microalbuminuria. The generalizability of the present findings to larger populations with or without diabetes is hence unknown. Furthermore, the inflammatory parameters measured were limited to hsCRP and U-LTE4. Notably, no salivary samples were available in this cohort for measures of S-LTB4, which could provide further information for a potential association of the leukotriene pathway with endothelial dysfunction . Finally, the lack of association of U-LTE4 with either endothelial function or anthropometric should be interpreted with care given the major confounder of renal function in the present cohort.
The present results did not establish an association of either of the two studied inflammatory disease markers, CRP and U-LTE4, with endothelial dysfunction in patients with type 2 diabetes. However, eGFR was an independent predictor of U-LTE4, but not CRP, in this patient cohort. In conclusion, these results indicate that renal function should be taken into consideration when studying U-LTE4 as biomarker. Since leukotrienes are undetectable in plasma [40, 41], these findings suggest that other means of leukotriene measures should be considered in such patients groups, such as salivary LTB  or assessment of leukotriene synthesis from ex vivo stimulated leukocytes .
Conflict of Interests
The authors declare that there is no conflict of interests.
This study was supported by the Swedish Research Council (Grants nos. 2011-2988 and 10857) and the Swedish Heart and Lung Foundation (Grants nos. 20120474, 20120827, and 20100259).
- M. Bäck, S.-E. Dahlén, J. M. Drazen et al., “International union of basic and clinical pharmacology. LXXXIV: leukotriene receptor nomenclature, distribution, and pathophysiological functions,” Pharmacological Reviews, vol. 63, no. 3, pp. 539–584, 2011.
- F. Stanke-Labesque, M. Bäck, B. Lefebvre et al., “Increased urinary leukotriene E4 excretion in obstructive sleep apnea: effects of obesity and hypoxia,” Journal of Allergy and Clinical Immunology, vol. 124, no. 2, pp. 364.e2–370.e2, 2009.
- M. Carry, V. Korley, J. T. Willerson, L. Weigelt, A. W. Ford-Hutchinson, and P. Tagari, “Increased urinary leukotriene excretion in patients with cardiac ischemia: in vivo evidence for 5-lipoxygenase activation,” Circulation, vol. 85, no. 1, pp. 230–236, 1992.
- G. Hardy, R. Boizel, J. Bessard et al., “Urinary leukotriene E4 excretion is increased in type 1 diabetic patients: a quantification by liquid chromatography-tandem mass spectrometry,” Prostaglandins and Other Lipid Mediators, vol. 78, no. 1–4, pp. 291–299, 2005.
- M. Bäck, A. Sultan, O. Ovchinnikova, and G. K. Hansson, “5-Lipoxygenase-activating protein: a potential link between innate and adaptive immunity in atherosclerosis and adipose tissue inflammation,” Circulation Research, vol. 100, no. 7, pp. 946–949, 2007.
- M. Spite, J. Hellmann, Y. Tang et al., “Deficiency of the leukotriene B4 receptor, BLT-1, protects against systemic insulin resistance in diet-induced obesity,” Journal of Immunology, vol. 187, no. 4, pp. 1942–1949, 2011.
- I. Mothe-Satney, C. Filloux, H. Amghar et al., “Adipocytes secrete leukotrienes: contribution to obesity-associated inflammation and insulin resistance in mice,” Diabetes, vol. 61, pp. 2311–2319, 2012.
- F. Paneni, J. A. Beckman, M. A. Creager, and F. Cosentino, “Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: part I,” European Heart Journal, vol. 34, no. 31, pp. 2436–2443, 2013.
- A. E. Caballero, S. Arora, R. Saouaf et al., “Microvascular and macrovascular reactivity is reduced in subjects at risk for type 2 diabetes,” Diabetes, vol. 48, no. 9, pp. 1856–1862, 1999.
- M. Bäck, “Inhibitors of the 5-lipoxygenase pathway in atherosclerosis,” Current Pharmaceutical Design, vol. 15, no. 27, pp. 3116–3132, 2009.
- G. Riccioni and M. Bäck, “Leukotrienes as modifiers of preclinical atherosclerosis?” The Scientific World Journal, vol. 2012, Article ID 490968, 2012.
- M. Bäck, “Leukotriene receptors: crucial components in vascular inflammation,” The Scientific World Journal, vol. 7, pp. 1422–1439, 2007.
- J. L. Ortiz, I. Gorenne, J. Cortijo et al., “Leukotriene receptors on human pulmonary vascular endothelium,” British Journal of Pharmacology, vol. 115, no. 8, pp. 1382–1386, 1995.
- M. Bäck, X. Norel, L. Walch et al., “Prostacyclin modulation of contractions of the human pulmonary artery by cysteinyl-leukotrienes,” European Journal of Pharmacology, vol. 401, no. 3, pp. 389–395, 2000.
- C. Labat, M. Temmar, E. Nagy et al., “Inflammatory mediators in saliva associated with arterial stiffness and subclinical atherosclerosis,” Journal of Hypertension, 2013.
- C. M. McEniery, S. Wallace, I. S. MacKenzie et al., “Endothelial function is associated with pulse pressure, pulse wave velocity, and augmentation index in healthy humans,” Hypertension, vol. 48, no. 4, pp. 602–608, 2006.
- A. Rafnsson, F. Böhm, M. Settergren, A. Gonon, K. Brismar, and J. Pernow, “The endothelin receptor antagonist bosentan improves peripheral endothelial function in patients with type 2 diabetes mellitus and microalbuminuria: a randomised trial,” Diabetologia, vol. 55, no. 3, pp. 600–607, 2012.
- A. Nohria, M. Gerhard-Herman, M. A. Creager, S. Hurley, D. Mitra, and P. Ganz, “Role of nitric oxide in the regulation of digital pulse volume amplitude in humans,” Journal of Applied Physiology, vol. 101, no. 2, pp. 545–548, 2006.
- P. O. Bonetti, G. W. Barsness, P. C. Keelan et al., “Enhanced external counterpulsation improves endothelial function in patients with symptomatic coronary artery disease,” Journal of the American College of Cardiology, vol. 41, no. 10, pp. 1761–1768, 2003.
- N. M. Hamburg and E. J. Benjamin, “Assessment of endothelial function using digital pulse amplitude tonometry,” Trends in Cardiovascular Medicine, vol. 19, no. 1, pp. 6–11, 2009.
- M. Settergren, F. Böhm, L. Rydén, and J. Pernow, “Cholesterol lowering is more important than pleiotropic effects of statins for endothelial function in patients with dysglycaemia and coronary artery disease,” European Heart Journal, vol. 29, no. 14, pp. 1753–1760, 2008.
- L. Ghiadoni, A. Cupisti, Y. Huang et al., “Endothelial dysfunction and oxidative stress in chronic renal failure,” Journal of Nephrology, vol. 17, no. 4, pp. 512–519, 2004.
- F. U. S. Mattace-Raso, T. J. M. van der Cammen, I. M. van der Meer et al., “C-reactive protein and arterial stiffness in older adults: the Rotterdam Study,” Atherosclerosis, vol. 176, no. 1, pp. 111–116, 2004.
- S. Makita, M. Nakamura, and K. Hiramori, “The association of C-reactive protein levels with carotid intima-media complex thickness and plaque formation in the general population,” Stroke, vol. 36, no. 10, pp. 2138–2142, 2005.
- H. Hashimoto, K. Kitagawa, H. Hougaku et al., “C-reactive protein is an independent predictor of the rate of increase in early carotid atherosclerosis,” Circulation, vol. 104, no. 1, pp. 63–67, 2001.
- P. M. Ridker, J. E. Buring, N. R. Cook, and N. Rifai, “C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: an 8-year follow-up of 14 719 initially healthy American women,” Circulation, vol. 107, no. 3, pp. 391–397, 2003.
- O. Schnell, I. Amann-Zalan, Z. Jelsovsky et al., “Changes in A1C levels are significantly associated with changes in levels of the cardiovascular risk biomarker hs-CRP: results from the SteP study,” Diabetes Care, vol. 36, no. 7, pp. 2084–2089, 2013.
- K. S. Heffernan, R. H. Karas, E. A. Patvardhan, H. Jafri, and J. T. Kuvin, “Peripheral arterial tonometry for risk stratification in men with coronary artery disease,” Clinical Cardiology, vol. 33, no. 2, pp. 94–98, 2010.
- C. Vitale, E. Cerquetani, M. Wajngarten et al., “In patients with coronary artery disease endothelial function is associated with plasma levels of C-reactive protein and is improved by optimal medical therapy,” Italian Heart Journal, vol. 4, no. 9, pp. 627–632, 2003.
- S. Verma, C.-H. Wang, E. Lonn et al., “Cross-sectional evaluation of brachial artery flow-mediated vasodilation and C-reactive protein in healthy individuals,” European Heart Journal, vol. 25, no. 19, pp. 1754–1760, 2004.
- P. Giouleka, G. Papatheodorou, P. Lyberopoulos et al., “Body mass index is associated with leukotriene inflammation in asthmatics,” European Journal of Clinical Investigation, vol. 41, no. 1, pp. 30–38, 2011.
- R. Boizel, G. Bruttmann, P. Y. Benhamou, S. Halimi, and F. Stanke-Labesque, “Regulation of oxidative stress and inflammation by glycaemic control: evidence for reversible activation of the 5-lipoxygenase pathway in type 1, but not in type 2 diabetes,” Diabetologia, vol. 53, no. 9, pp. 2068–2070, 2010.
- N. Higashi, M. Taniguchi, H. Mita et al., “Clinical features of asthmatic patients with increased urinary leukotriene E4 excretion (hyperleukotrienuria): involvement of chronic hyperplastic rhinosinusitis with nasal polyposis,” The Journal of Allergy and Clinical Immunology, vol. 113, no. 2, pp. 277–283, 2004.
- S. W. Reinhold, T. Scherl, B. Stolcker et al., “Lipoxygenase products in the urine correlate with renal function and body temperature but not with acute transplant rejection,” Lipids, vol. 48, no. 2, pp. 167–175, 2013.
- R. Petric and A. W. Ford-Hutchinson, “Elevated cysteinyl leukotriene excretion in experimental glomerulonephritis,” Kidney International, vol. 46, no. 5, pp. 1322–1329, 1994.
- J. M. Valdivielso, A. Montero, K. F. Badr, and K. A. Munger, “Inhibition of 5-lipoxygenase activating protein decreases proteinuria in diabetic rats,” Journal of nephrology, vol. 16, no. 1, pp. 85–94, 2003.
- E. Noiri, T. Yokomizo, A. Nakao et al., “An in vivo approach showing the chemotactic activity of leukotriene B4 in acute renal ischemic-reperfusion injury,” Proceedings of the National Academy of Sciences of the United States of America, vol. 97, no. 2, pp. 823–828, 2000.
- D. W. Butterly, R. F. Spurney, P. Ruiz, R. Griffiths, C. Albrightson, and T. M. Coffman, “A role for leukotrienes in cyclosporine nephrotoxicity,” Kidney International, vol. 57, no. 6, pp. 2586–2593, 2000.
- K. P. Moore, G. W. Taylor, N. H. Maltby et al., “Increased production of cysteinyl leukotrienes in hepatorenal syndrome,” Journal of Hepatology, vol. 11, no. 2, pp. 263–271, 1990.
- D. A. Brezinski, R. W. Nesto, and C. N. Serhan, “Angioplasty triggers intracoronary leukotrienes and lipoxin A4: impact of aspirin therapy,” Circulation, vol. 86, no. 1, pp. 56–63, 1992.
- X. Houard, V. Ollivier, L. Louedec, J.-B. Michel, and M. Bäck, “Differential inflammatory activity across human abdominal aortic aneurysms reveals neutrophil-derived leukotriene B4 as a major chemotactic factor released from the intraluminal thrombus,” FASEB Journal, vol. 23, no. 5, pp. 1376–1383, 2009.
- F. Stanke-Labesque, J. L. Pepin, T. de Jouvencel et al., “Leukotriene B4 pathway activation and atherosclerosis in obstructive sleep apnea,” Journal of Lipid Research, vol. 53, pp. 1944–1951, 2012.
Copyright © 2013 Arnar Rafnsson and Magnus Bäck. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.