This article has Open Peer Review reports available.
Chest pain in primary care: is the localization of pain diagnostically helpful in the critical evaluation of patients? - A cross sectional study
© Bösner et al.; licensee BioMed Central Ltd. 2013
Received: 28 May 2013
Accepted: 25 September 2013
Published: 18 October 2013
Chest pain is a common complaint and reason for consultation in primary care. Traditional textbooks still assign pain localization a certain discriminative role in the differential diagnosis of chest pain. The aim of our study was to synthesize pain drawings from a large sample of chest pain patients and to examine whether pain localizations differ for different underlying etiologies.
We conducted a cross-sectional study including 1212 consecutive patients with chest pain recruited in 74 primary care offices in Germany. Primary care providers (PCPs) marked pain localization and radiation of each patient on a pictogram. After 6 months, an independent interdisciplinary reference panel reviewed clinical data of every patient, deciding on the etiology of chest pain at the time of patient recruitment. PCP drawings were entered in a specially designed computer program to produce merged pain charts for different etiologies. Dissimilarities between individual pain localizations and differences on the level of diagnostic groups were analyzed using the Hausdorff distance and the C-index.
Pain location in patients with coronary heart disease (CHD) did not differ from the combined group of all other patients, including patients with chest wall syndrome (CWS), gastro-esophageal reflux disease (GERD) or psychogenic chest pain. There was also no difference in chest pain location between male and female CHD patients.
Pain localization is not helpful in discriminating CHD from other common chest pain etiologies.
Patients with chest pain are encountered on a regular basis in primary care. In different studies the incidence of chest pain varies according to setting, country, and inclusion criteria [1–3]. There is a wide range of different underlying diseases including coronary heart disease (CHD) [4, 5].
In regard to the diagnostic work-up of chest pain patients primary care providers (PCPs) are trained to elicit, among other information, the exact pain location. Both clinical guidelines and standard textbooks recommend a detailed clinical history including pain location and radiation [6, 7]. Several diagnostic studies and meta-analyses have examined the diagnostic value of pain location mainly in regard to CHD and acute coronary syndrome (ACS) [8–12]. In these studies, pain locations were normally marked on a pictogram either by patients or the attending physician. Marked areas were consequently aggregated (e.g. 'upper left pain’) for analysis, resulting in a loss of detailed data [9, 10, 13]. Most of these investigations were performed in secondary care settings and results are setting-specific and inconsistent.
Pain maps have been used frequently in other areas of research like low back pain [14, 15], migraine headaches , or temporomandibular disorders and fibromyalgia syndrome . While some of these studies still used conventional grid methods as the above quoted chest pain studies did , other authors applied advanced methodological techniques superimposing pain drawings  and transforming data into two-dimensional color coded images .
In our study we applied a newly developed technique to analyze pain drawings of a large cohort of unselected and consecutively recruited primary care patients with chest pain in order to find out whether pain localization is helpful to discriminate between CHD and other diseases.
The primary aim of our original cross-sectional diagnostic study was to investigate the diagnostic accuracy of signs and symptoms for chest pain patients with CHD . A detailed account of our study design can be found there. In this article we report results of a sub-analysis with regard to the diagnostic value of pain location in patients with chest pain.
Participating PCPs and patients
Out of 209 contacted PCPs, 74 (35.4%) agreed to participate in the study. PCPs consecutively recruited every patient above 35 years with pain localized in the area between the clavicles and the lower costal margins, and the anterior to the posterior axillary lines. Patients were eligible irrespective of the acute or chronic nature of their complaints, including known conditions like CHD, and were also recruited during home visits and emergency calls. Patients were excluded if their chest pain had subsided for more than one month, had already been investigated, or in case of a follow-up visit for previously diagnosed chest pain. The study protocol was consistent with the Declaration of Helsinki and all participants gave their informed consent.
Data collection and analysis
Study assistants contacted all patients by phone both six weeks and six months after the initial consultation and asked about the course of the patient’s chest pain and treatment thereof. Participating PCPs requested discharge letters from specialists and hospitals.
Precautions against selection bias
We emphasized to all PCPs the importance of recruiting every patient with chest pain and visited PCP practices at four week intervals to check compliance with study procedures. In addition, we performed random audits to identify cases of chest pain not included in the study.
Diagnosis and reference standard
A reference panel including 1 cardiologist, 1 PCP, and 1 research associate from our department reviewed baseline and follow-up data of each patient. They discussed and decided on the most likely etiology of the individual patient’s chest pain at the time of the index test (delayed type reference standard). The PCP’s initial diagnosis was considered among other clinical data.
Digitalization of pain mapping information
To be able to perform computerized analysis on the pain region data, a computer application was created which allowed a research assistant do draw each patient’s pain regions, as well as the arrows marking radiation directions, into an exact digital replica of the pictogram on the report forms using the mouse pointer. The regions are captured in binary (black and white) images with black pixels representing a region with pain. The data generated for each patient consists of two binary images with a resolution of 900×516 pixels, one containing the pain regions, the other one the pain radiation arrows.
Computing process and image calculations
We superimposed all our patients’ images with the respective pain regions of a group representing the number of overlaps at each pixel position. The color of the resulting graphs reflected the degree of overlap. A large number of overlaps are represented in red colors and a small number of overlaps in blue colors. In order to be able to compare pain regions across different images, the color range was scaled to the total number of patients in the group. This means that only a region where all patients of this group overlap will be represented with the maximum possible color value (red).
Statistical analysis of differences between pain maps
In order to answer the original question concerning a possible distinction between diagnostic groups on the basis of their respective pain regions, individual pain localizations were compared in terms of a suitable measure of dissimilarity. Differences on the level of diagnostic groups were then analyzed on the basis of the pair-wise dissimilarity degrees thus produced.
More specifically, dissimilarities between individual pain localizations were measured in terms of the Hausdorff distance, which is a well-known and widely used measure for the distance of subsets of a metric space .
The comparison of two groups of patients with different diagnosis was accomplished by means of the C-index . This index compares the pair-wise inter-group dissimilarities (i.e., the dissimilarity between two patients from the same group) with the pair wise intra-group dissimilarities (i.e., the dissimilarity between two patients from different groups). It ranges between 0 and 1 and assumes values close to 0 if the inter-group dissimilarities are large compared to the intra-group dissimilarities, thus indicating that the two groups can be well separated. A value of 0.5 indicates equal inter-group and intra-group dissimilarities and values close to 1 point to higher intra- than inter-group dissimilarities.
Finally, the significance of the C-index computed was determined by means of a standard permutation test. This test delivers a p-value which corresponds to the probability to obtain a smaller C-index if the assignment of the patients to the two groups is permuted in a random way.
The whole study was approved by the Ethics Committee of the Faculty of Medicine, University of Marburg. The study complies with the declaration of Helsinki.
PCPs and patients characteristics
Final diagnoses in patients presenting with chest pain to their GP (n = 1212)
Frequency (n = 1212)
Chest wall syndrome*
Coronary Heart Disease
Upper respiratory infections
Gastroesophageal reflux disease
Benign stomach problems
Pain localization: CHD vs. other diseases
Comparison of different pain regions: Hausdorff-distance based clustering results
CHD vs. all other chest pain etiologies
CHD vs. chest wall syndrome (CWS)
CHD vs. GERD
CHD vs. psychogenic chest pain
CHD (male patients) vs. CHD (female patients)
CWS patients who assume a cardiac origin of their pain vs. CWS patients who do not assume a cardiac origin
Pain localization: CHD by gender
Pain localization: chest wall syndrome and patient assumptions
We examined in a large prospective primary care study whether pain localization is helpful to discriminate between CHD and other diseases in chest pain patients. Pain localizations of all major chest pain etiologies (CHD, CWS, GERD and psychogenic chest pain) were mainly situated on the left anterior chest and did not help to discriminate between CHD and other diseases.
Strengths of our study are a large primary care based consecutive sample which is highly representative, the prospective design and low drop-out rates during the follow up period. Study procedures such as random audits reduced the possibility of selection bias. An interdisciplinary team of PCPs and cardiologists provided a precise diagnosis as reference standard. As we did not cluster pain localization data but plotted the original drawings with the help of a specially designed computer program we could maintain highest data integrity for graphical and statistical analysis.
As we did not interfere with the work-up provided by participating PCPs, for some patients only limited clinical data were available to the reference panel. Since data from the original questionnaire, also including PCPs’ provisional diagnoses, were also used by the panel for decision making, there may be a degree of incorporation bias in regard to the final diagnoses .
We did not find pain localization helpful in discriminating between CHD and other diseases. This stands in contrast to Gencer et al. who analyzed a sample of 672 chest pain patients in primary care and found an association between substernal pain and CHD . Several other studies could find no or only limited use of the localization of chest pain in predicting which patients would eventually have ACS or acute myocardial infarction (AMI) [9, 23–27]. Cooke et al. examined chest pain characteristics in a highly selected patient population with chronic stable CHD and could also find no differences in pain localization . Only one study in patients referred for coronary angiography found that pain to the left of the sternum occurs more frequently in patients with normal coronary arteries . In a systematic review modeling the investigation of acute chest pain in primary care conducted by Mant et al., localization of chest pain was not helpful in ruling ACS in or out .
Chest wall syndrome (CWS) constitutes the most common etiology of chest pain in primary care [30, 31]. The question whether pain localization is helpful to distinguish between CWS and CHD is therefore of high practical relevance for PCPs. From a pathophysiological point of view one would expect that pain localization of a higher number of patients with CWS would be more or less equally distributed with no preference for one side of the thorax.
However, our data show that CWS, like CHD, is mainly situated on the left anterior chest side, and that location does not discriminate between these two diseases. Our findings are supported by Verdon et al. who observed in a cohort of 300 primary care patients with CWS also the main pain localization on the left or median-left part of the chest wall . Wise et al. describe in a selected population of 100 patients with negative coronary arteriography 69 patients with chest wall tenderness, most commonly situated in the sternal and left anterior chest wall area . One has to assume that even in an 'unselected’ primary care population as observed in our study as well as by Verdon et al., there is already a certain selection effect. Due to public health campaigns the general population associates left thoracic pain mainly with the danger of CHD and will most likely contact a doctor more frequently than if the same pain occurred in any other thoracic area. This is supported by our sub-analysis presented in Figure 7 which compares CWS patients who assume a cardiac origin of their pain with CWS patients without this assumption. The first group shows a statistically significant more clustered pain distribution where the heart is situated.
Symptoms of GERD are a common complaint in primary care patients . The pain caused by GERD can mimic the pain caused by CHD. In a study conducted by Davies et al. classical features of angina pectoris were equally common in CHD patients and patients with esophageal disease . As both organs are situated near to each other and the resulting pain is in each case of visceral nature, one would expect few differences in pain localization, which is also supported by our data.
Psychogenic chest pain ranks among the 5 most frequent etiologies of chest pain in primary care [30, 31]. Beside panic disorders, anxiety and depression prevail in these patients . As these patients themselves often assume a cardiac origin they are in particular danger of receiving unnecessary further investigations. Pain location is classically described as uncharacteristic affecting multiple sites of the chest and being difficult to distinguish from CHD . Our findings show nearly a complete overlap of pain regions in patients with psychogenic chest pain compared to CHD induced chest pain. On the one hand we would postulate similar self-selection mechanisms as already described above for CWS patients. Additionally it might be the very nature of psychogenic pain to be more projected towards the cardiac area. It could also be shown that myocardial infarction patients report left sided chest pain during their prodromal phase in the same frequency as a control group of patients with hyperventilation and/or functional complaints .
Finally we analyzed pain distribution in male and female CHD patients and could also find no difference. There are many studies that have examined gender differences in symptom presentation, mainly for ACS or AMI [38–40]. All of these have been performed in emergency departments and the authors did not specifically investigate left sided chest pain, but other pain regions. Major differences in pain distribution were not described.
In summary our results show that chest pain localization is neither helpful in discriminating CHD patients from other patients nor is it helpful to identify or exclude other chest pain etiologies. While in some diseases there might be an a priori high overlap of pain localization (e.g. CHD and GERD), in other conditions (like CWS or psychogenic chest pain) pain localization might trigger consultation of a health care provider. Consequently, the diagnostic value of pain localization in this instance is already 'used up’ and no longer of diagnostic value in the primary care setting. Similar phenomena of self-selection bias have been described for other clinical settings [41, 42] and may have contributed to our findings.
In contrast to the information still provided in many medical textbooks pain localization is not helpful in discriminating CHD from other common chest pain etiologies. Doctors should focus more on other clinical characteristics when evaluating chest pain patients .
The original study was funded by Federal Ministry of Education and Research (BMBF — grant no FKZ 01GK0401). The funding source had no involvement in the study.
- Svavarsdottir AE, Jonasson MR, Gudmundsson GH, Fjeldsted K: Chest pain in family practice. Diagnosis and long-term outcome in a community setting. Can Fam Physician. 1996, 42: 1122-1128.PubMedGoogle Scholar
- Nilsson S, Scheike M, Engblom D, Karlsson LG, Molstad S, Akerlind I, Ortoft K, Nylander E: Chest pain and ischaemic heart disease in primary care. BrJ GenPract. 2003, 53 (490): 378-382.Google Scholar
- Verdon F, Burnand B, Herzig L, Junod M, Pecoud A, Favrat B: Chest wall syndrome among primary care patients: a cohort study. BMC Fam Pract. 2007, 12 (8): 51-View ArticleGoogle Scholar
- Buntinx F, Knockaert D, Bruyninckx R, De Blaey N, Aerts M, Knottnerus JA, Delooz H: Chest pain in general practice or in the hospital emergency department: is it the same?. Fam Pract. 2001, 18 (6): 586-589. 10.1093/fampra/18.6.586.View ArticlePubMedGoogle Scholar
- Klinkman MS, Stevens D, Gorenflo DW: Episodes of care for chest pain: a preliminary report from MIRNET. Michigan Research Network. J FamPract. 1994, 38 (4): 345-352.Google Scholar
- Smeeth L, Skinner JS, Ashcroft J, Hemingway H, Timmis A: NICE clinical guideline: chest pain of recent onset. Br J Gen Pract. 2010, 60 (577): 607-610. 10.3399/bjgp10X515124.View ArticlePubMedPubMed CentralGoogle Scholar
- Hurst JW, Morris DC: Chest pain. 2001, Armonk N.Y: Futura PubGoogle Scholar
- Short D: Diagnosis of slight and subacute coronary attacks in the community. Br Heart J. 1981, 45 (3): 299-310. 10.1136/hrt.45.3.299.View ArticlePubMedPubMed CentralGoogle Scholar
- Everts B, Karlson BW, Wahrborg P, Hedner T, Herlitz J: Localization of pain in suspected acute myocardial infarction in relation to final diagnosis, age and sex, and site and type of infarction. Heart Lung. 1996, 25 (6): 430-437. 10.1016/S0147-9563(96)80043-4.View ArticlePubMedGoogle Scholar
- Berger JP, Buclin T, Haller E, Van Melle G, Yersin B: Right arm involvement and pain extension can help to differentiate coronary diseases from chest pain of other origin: a prospective emergency ward study of 278 consecutive patients admitted for chest pain. J Intern Med. 1990, 227 (3): 165-172. 10.1111/j.1365-2796.1990.tb00138.x.View ArticlePubMedGoogle Scholar
- Mant J, McManus RJ, Oakes RA, Delaney BC, Barton PM, Deeks JJ, Hammersley L, Davies RC, Davies MK, Hobbs FD: Systematic review and modelling of the investigation of acute and chronic chest pain presenting in primary care. Health Technol Assess. 2004, 8 (2): iii, 1-iii158.View ArticleGoogle Scholar
- Chun AA, McGee SR: Bedside diagnosis of coronary artery disease: a systematic review. Am J Med. 2004, 117 (5): 334-343. 10.1016/j.amjmed.2004.03.021.View ArticlePubMedGoogle Scholar
- Wu EB, Hodson F, Chambers JB: A simple score for predicting coronary artery disease in patients with chest pain. QJM. 2005, 98 (11): 803-811. 10.1093/qjmed/hci122.View ArticlePubMedGoogle Scholar
- Mann NH, Brown MD, Hertz DB, Enger I, Tompkins J: Initial-impression diagnosis using low-back pain patient pain drawings. Spine. 1993, 18 (1): 41-53. 10.1097/00007632-199301000-00008.View ArticlePubMedGoogle Scholar
- Takata K, Hirotani H: Pain drawing in the evaluation of low back pain. Int Orthop. 1995, 19 (6): 361-366.View ArticlePubMedGoogle Scholar
- Russell MB, Iversen HK, Olesen J: Improved description of the migraine aura by a diagnostic aura diary. Cephalalgia. 1994, 14 (2): 107-117. 10.1046/j.1468-2982.1994.1402107.x.View ArticlePubMedGoogle Scholar
- Pfau DB, Rolke R, Nickel R, Treede RD, Daublaender M: Somatosensory profiles in subgroups of patients with myogenic temporomandibular disorders and Fibromyalgia Syndrome. Pain. 2009, 147 (1–3): 72-83.View ArticlePubMedGoogle Scholar
- Bösner S, Becker A, Abu Hani M, Keller H, Sonnichsen AC, Haasenritter J, Karatolios K, Schaefer JR, Baum E, Donner-Banzhoff N: Accuracy of symptoms and signs for coronary heart disease assessed in primary care. Br J Gen Pract. 2010, 60 (575): 246-257. 10.3399/bjgp10X502137.View ArticleGoogle Scholar
- Huttenlocher DP, Klanderman GA, Rucklidge WJ: Comparing images using the Hausdorff distance. IEEE Trans Pattern Anal Mach Intell. 1993, 15 (9): 850-863. 10.1109/34.232073.View ArticleGoogle Scholar
- Hubert L, Schultz J: Quadratic assignment as a general data-analysis strategy. Br J Math Stat Psych. 1976, 29: 190-241. 10.1111/j.2044-8317.1976.tb00714.x.View ArticleGoogle Scholar
- Whiting P, Rutjes AW, Reitsma JB, Glas AS, Bossuyt PM, Kleijnen J: Sources of variation and bias in studies of diagnostic accuracy: a systematic review. Ann Intern Med. 2004, 140 (3): 189-202. 10.7326/0003-4819-140-3-200402030-00010.View ArticlePubMedGoogle Scholar
- Gencer B, Vaucher P, Herzig L, Verdon F, Ruffieux C, Boesner S, Burnand B, Bischoff T, Donner-Banzhoff N, Favrat B: Ruling out coronary heart disease in primary care patients with chest pain: a clinical prediction score. BMC Med. 2010, 8 (1): 9-10.1186/1741-7015-8-9.View ArticlePubMedPubMed CentralGoogle Scholar
- Sawe U: Pain in acute myocardial infarction. A study of 137 patients in a coronary care unit. Acta Med Scand. 1971, 190 (1–2): 79-81.PubMedGoogle Scholar
- Bulgiba AM, Razaz M: How well can signs and symptoms predict AMI in the Malaysian population?. Int J Cardiol. 2005, 102 (1): 87-93. 10.1016/j.ijcard.2004.04.002.View ArticlePubMedGoogle Scholar
- Logan RL, Wong F, Barclay J: Symptoms Associated with Myocardial-Infarction - Are They of Diagnostic-Value. New Zeal Med J. 1986, 99 (800): 276-278.PubMedGoogle Scholar
- Grijseels EW, Deckers JW, Hoes AW, Hartman JA, Der DE V, Van Loenen E, Simoons ML: Pre-hospital triage of patients with suspected myocardial infarction. Evaluation of previously developed algorithms and new proposals. Eur Heart J. 1995, 16 (3): 325-332.PubMedGoogle Scholar
- Goodacre SW, Angelini K, Arnold J, Revill S, Morris F: Clinical predictors of acute coronary syndromes in patients with undifferentiated chest pain. QJM. 2003, 96 (12): 893-898. 10.1093/qjmed/hcg152.View ArticlePubMedGoogle Scholar
- Cooke RA, Smeeton N, Chambers JB: Comparative study of chest pain characteristics in patients with normal and abnormal coronary angiograms. Heart. 1997, 78 (2): 142-146.View ArticlePubMedPubMed CentralGoogle Scholar
- Mukerji V, Alpert MA, Hewett JE, Parker BM: Can Patients with Chest Pain and Normal Coronary-Arteries Be Discriminated from Those with Coronary-Artery Disease Prior to Coronary Angiography. Angiology. 1989, 40 (4): 276-282. 10.1177/000331978904000406.View ArticlePubMedGoogle Scholar
- Bösner S, Becker A, Haasenritter J, Abu Hani M, Keller H, Sönnichsen AC, Karatolios K, Schaefer JR, Seitz G, Baum E, et al: Chest pain in primary care: Epidemiology and pre-work-up probabilities. Eur J Gen Pract. 2009, 15 (3): 141-146. 10.3109/13814780903329528.View ArticlePubMedGoogle Scholar
- Verdon F, Herzig L, Burnand B, Bischoff T, Pecoud A, Junod M, Muhlemann N, Favrat B: Chest pain in daily practice: occurrence, causes and management. Swiss Med Wkly. 2008, 138 (23–24): 340-347.PubMedGoogle Scholar
- Wise CM, Semble EL, Dalton CB: Musculoskeletal chest wall syndromes in patients with noncardiac chest pain: a study of 100 patients. Arch Phys Med Rehabil. 1992, 73 (2): 147-149.PubMedGoogle Scholar
- Bruley Des Varannes S, Marek L, Humeau B, Lecasble M, Colin R: Gastroesophageal reflux disease in primary care. Prevalence, epidemiology and Quality of Life of patients. Gastroenterol Clin Biol. 2006, 30 (3): 364-370. 10.1016/S0399-8320(06)73189-X.View ArticlePubMedGoogle Scholar
- Davies HA, Jones DB, Rhodes J, Newcombe RG: Angina-like esophageal pain: differentiation from cardiac pain by history. J Clin Gastroenterol. 1985, 7 (6): 477-481. 10.1097/00004836-198512000-00007.View ArticlePubMedGoogle Scholar
- Katon W, Hall ML, Russo J, Cormier L, Hollifield M, Vitaliano PP, Beitman BD: Chest pain: relationship of psychiatric illness to coronary arteriographic results. Am J Med. 1988, 84 (1): 1-9.View ArticlePubMedGoogle Scholar
- Frankel B: “Chest Pain” in patients with anxiety disorders. Chest Pain. Edited by: Hurst JW. 2001, Morris DC. New York: Futura Publishing Company, Inc, 415-428.Google Scholar
- Beunderman R, Duyvis DJ: Myocardial-Infarction Patients during the Prodromal and Acute Phase - a Comparison with Patients with a Diagnosis of Noncardiac Chest Pain. Psychother Psychosom. 1983, 40 (1–4): 129-136.View ArticlePubMedGoogle Scholar
- Arslanian-Engoren C, Patel A, Fang J, Armstrong D, Kline-Rogers E, Duvernoy CS, Eagle KA: Symptoms of men and women presenting with acute coronary syndromes. Am J Cardiol. 2006, 98 (9): 1177-1181. 10.1016/j.amjcard.2006.05.049.View ArticlePubMedGoogle Scholar
- Meischke H, Larsen MP, Eisenberg MS: Gender differences in reported symptoms for acute myocardial infarction: impact on prehospital delay time interval. Am J Emerg Med. 1998, 16 (4): 363-366. 10.1016/S0735-6757(98)90128-0.View ArticlePubMedGoogle Scholar
- DeVon HA, Zerwic JJ: Symptoms of acute coronary syndromes: are there gender differences? A review of the literature. Heart Lung. 2002, 31 (4): 235-245. 10.1067/mhl.2002.126105.View ArticlePubMedGoogle Scholar
- Nilsen RM, Vollset SE, Gjessing HK, Skjaerven R, Melve KK, Schreuder P, Alsaker ER, Haug K, Daltveit AK, Magnus P: Self-selection and bias in a large prospective pregnancy cohort in Norway. Paediatr Perinat Epidemiol. 2009, 23 (6): 597-608. 10.1111/j.1365-3016.2009.01062.x.View ArticlePubMedGoogle Scholar
- Lieu JE, Dewan K: Assessment of self-selection bias in a pediatric unilateral hearing loss study. Otolaryngol Head Neck Surg. 2010, 142 (3): 427-433. 10.1016/j.otohns.2009.11.035.View ArticlePubMedPubMed CentralGoogle Scholar
- Bösner S, Haasenritter J, Becker A, Karatolios K, Vaucher P, Gencer B, Herzig L, Heinzel-Gutenbrunner M, Schaefer JR, Abu Hani M, et al: Ruling out coronary artery disease in primary care: development and validation of a simple prediction rule. CMAJ. 2010, 182 (12): 1295-1300. 10.1503/cmaj.100212.View ArticlePubMedPubMed CentralGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2296/14/154/prepub
This article is published under license to 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.