Malaria transmission in non-endemic areas: case report, review of the literature and implications for public health management
© Zoller et al; licensee BioMed Central Ltd. 2009
Received: 26 November 2008
Accepted: 20 April 2009
Published: 20 April 2009
In non-endemic areas, malaria is rare and locally acquired infections, particularly with Plasmodium falciparum, are exceptional events. The diagnosis is, therefore, likely to be delayed or missed in patients without a relevant travel history. This report describes a case of falciparum malaria in Berlin, Germany, in a patient who had not been to an endemic area for more than a decade. Potential routes of vector-related and direct transmission were evaluated, particularly with regard to a possible danger to the public. A review of the literature was conducted regarding possible routes of transmission and their probability assessed. Genotyping of parasite isolates of this and another patient with malaria admitted 16 days before revealed homology between the two strains. In a local entomological survey, anopheline vectors on the hospital grounds as well as in the residential area of both patients were found. Despite intensive investigations, the mode of transmission remained obscure. In this context, possible routes of vector-borne and direct occupational/accidental transmission in a major European city are reviewed and discussed, providing information and guidance in case other similar events occur elsewhere. Examples for investigations and measures to be taken in such a situation are provided. When local malaria transmission within a large non-immune population cannot be ruled out, genotyping of parasite isolates, local entomological surveys, preparedness for secondary cases, expert consultations in a multidisciplinary team and careful information management are essential. Malaria acquired in non-endemic areas remains an unlikely, but possible event for which awareness needs to be maintained.
Transmission of malaria in a non-endemic area is an extremely unusual event, but is possible under certain conditions. For those affected, the combination of a missed or delayed diagnosis along with the high fatality of Plasmodium falciparum infection may result in serious consequences. These rare occasions pose a challenge to local and public health authorities.
In the European Union, 6,000–10,000 imported cases of malaria and 26 to 78 attributable deaths have been reported annually between 1994 and 2006 . In sick travellers returning from malaria-endemic areas, a detailed travel history is mandatory and malaria must be excluded. Plasmodium falciparum- infected individuals without a relevant travel history are at high risk of a delayed/missed diagnosis and, consequently, of developing potentially life-threatening severe falciparum malaria.
Local transmission leading to isolated malaria cases in non-endemic regions has occurred under various circumstances including (i) "airport", (ii) "port", and (iii) "baggage" malaria, (iv) nosocomial transmission and (v) transmission by local competent vectors. Also, small malaria epidemics with continued localized transmission occurred in northern Germany and Berlin up to 1947 , and, more recently, in the USA .
With the aim to raise awareness for the potential of locally acquired malaria and to provide information and guidance, this report discusses possible routes of vector-borne and occupational/accidental transmission. As an example for investigations and measures to be taken in such a situation, a case of local transmission of P. falciparum in Berlin, Germany, which occurred in July 2007 is presented.
A consultant physician of the Department of Obstetrics & Gynaecology presented to the emergency room of the same university hospital in Berlin on 26 July 2007. She reported a four-day history of fever and headache. Common causes of fever were ruled out. On questioning, the patient stated that she had never been to Asia or Africa. However, she had travelled to Colombia eleven years before where she contracted malaria. Despite an extremely low probability of a tertian malaria relapse after this time, thick and thin blood films were examined, revealing Plasmodium falciparum at a density of 0.5% parasitized erythrocytes. Plasmodium falciparum mono-infection was confirmed by PCR . Other laboratory abnormalities included a thrombocytopaenia of 50/nL, an elevated C-reactive protein of 9.74 mg/dL, and leukocyturia. The patient was treated with oral artemether/lumefantrine and recovered fully within three days.
The consultant ("secondary case") encountered the primary case in person on only two occasions, once in the evening of admission to the labour ward (10 July) and then again the following morning. The secondary case, supervising the resident physicians, could not recall any direct physical contact on these two occasions, apart from possibly shaking hands. In particular, the consultant did not perform any medical procedures on the primary case at any time during the entire hospital stay, nor did she recall any needlestick injuries. On one occasion, the consultant hand-carried a fully bagged blood sample of the primary case, which had been collected by another physician on admission; she could, however, not recall seeing any blood on the outside of the bag.
Methods and results regarding possible modes of transmission and management
Suspecting local malaria transmission, all possible routes of transmission, as outlined above, had to be considered at the outset. One scenario included a possible danger to the public from infective mosquito(es) present on the hospital compound or in the residential area where both cases live.
Genetic homology of parasite strains
Parasite isolates from primary and secondary case were genotyped for the highly polymorphic P. falciparum msp-1 and msp-2 genes . The results revealed complete homology. This was confirmed in a second laboratory with a second set of samples and by extending genotyping to PfRRM microsatellite markers  and polymorphic DBL-alpha sequences . In contrast, human microsatellite markers showed dissimilarity between the two samples, excluding a mix-up or contamination of the samples taken from both cases .
Sources of occupational or accidental inoculation
The delivery room and maternity wards were thoroughly inspected and all persons directly or indirectly involved in the management of the primary case were repeatedly interviewed. No exposures compatible with a nosocomial transmission could be recalled. In particular, the absence of physical contact between index and secondary case was confirmed independently. The secondary case did not recall any injections, accidents or blood transfusions in the months preceding the malaria episode. She had no relevant skin damage during the possible direct-transmission window, apart from a small dressed wound on the wrist following a horsefly bite. Moreover, she stated that this wound had begun to close by the 10th July 2007. She could not recall changing the dressing whilst on the wards.
The index patient had used the same baggage both for her trip to the Ivory Coast and her hospital admission 11 days after returning. Both index patient and her husband stated that the bags had been completely emptied after returning to Berlin, and on inspection, nothing unusual was observed.
Contact or transmission in residential area
Further interviews with the two patients revealed that both live in the same, densely populated residential area of Berlin, approximately 1.8 km apart from each other. However, none of the places frequently visited by the patients in their daily activities were common to both.
In August 2007, an entomological survey on the hospital compound and the patients' area of residence was carried out. In both places, viable Anopheles larvae were found: on the hospital compound in small quantities in a fountain, and in the residential area in a backyard water butt. Morphological examination of the Anopheles larvae identified them as the species-complex Anopheles maculipennis s.l., which includes species with potential vectorial competence [8, 9]. Adult Anopheles mosquitoes were not found.
Information management of the suspected outbreak situation
Meanwhile, the federal health administration of Berlin convened a panel, which included infectious disease and public health specialists, along with communications officers. Although the occurrence of other mosquito-transmitted secondary cases seemed unlikely, this risk could not completely be ruled out at the time of the investigation. A selective approach to risk communication was used, with an alert sent only to hospitals in Berlin with emergency departments. The alert advised clinicians to consider malaria in cases of unexplained fever, particularly when coinciding with thrombocytopaenia. No further, presumably locally acquired, P. falciparum infections were reported to the health authorities. Public media did not report the incident.
Malaria has been endemic in Europe and particularly around the North Sea at least from the 7th century until the late 1950s . Martini  reported an epidemic on the German North Sea coast in 1826 with several thousand cases and many casualties; transmission was facilitated by unusually high summer temperatures. The last reported cases of autochthonous tertian malaria were observed in Berlin and Hamburg in 1947 . Plasmodium falciparum is thought to have never been endemic in Germany, but continues to be imported by travellers returning from endemic areas.
In the last six decades, only three cases of potentially local P. falciparum transmission have been reported from Germany. One of these, involving presumed transmission by local Anopheles plumbeus on a hospital compound in Duisburg, Western Germany, shows some parallels to the case reported here . Two simultaneous cases of supposed mosquito-borne infection occurred in 1994 close to a sewage plant in Berlin . In the present case, genetic analysis of parasite isolates did confirm the index patient as the source of the infection in the secondary patient.
Requirements for local mosquito transmission
Larvae of A. maculipennis s.l. were identified both on the hospital compound and in the patients' residential area. The time period required for completion of sporogony (extrinsic incubation time) is critical in determining the likelihood of mosquito-borne transmission. This period is highly temperature-dependent, but usually takes at least seven days under optimal average ambient temperatures in Africa of 35°C  and up to 30 days under ambient temperatures of 20°C . Mosquitoes of the A. maculipennis-species complex are potential vectors of P. falciparum .
Transmission by local Anopheles mosquitoes on the hospital compound
In Berlin, temperatures were exceptionally high exactly during the time of potential extrinsic incubation with night minimum temperatures >14°C in two nights and >17°C in the following five nights. Local temperature data was analysed and a probable extrinsic incubation time of 21 to 24 days calculated for P. falciparum . Mosquitoes resting within buildings (e.g. in the basement) may have been exposed to higher average temperatures. Adding to the extrinsic incubation period, the minimum intrinsic incubation time of P. falciparum infection in human experimental malaria is estimated as seven days . Assuming local transmission by anopheline vectors, the entire incubation period in this case (extrinsic and intrinsic) would amount to a minimum of 28 days, based on theoretical considerations using experimental data currently available. This, however, exceeds the timespan of 11 days between the index patient's admission and first symptoms in the secondary case (see Figure 1) arguing against a mosquito-borne transmission on the hospital grounds.
Transmission by local Anopheles in the residential area
The index and the secondary patient live in the same centrally located borough of Berlin, some 1.8 km apart. After returning from Ivory Coast on 30 June, the index patient stayed in this area for eleven days before being admitted to hospital. Theoretically, intrinsic plus extrinsic incubation period and the presence of An. maculipennis s.l. would allow a mosquito-borne transmission of P. falciparum in the city. However, it would be unusual, though not impossible, for a local anopheline species to travel more than 2 km within a short period of time. Furthermore, it seems very improbable that this unlikely event (of transmission in the residential area) should be followed by the even less probable coincidental meeting of index and secondary case in the same hospital as doctor and patient.
Eventually, genetic analysis showed a high degree of homology between both parasite isolates. Although this does not completely exclude a mosquito passage of P. falciparum (in which sexual parasite reproduction and genetic recombination occur), it renders both scenarios involving vector-borne transmission unlikely.
"Airport malaria" results from infective bites of mosquitoes, which travelled aboard an airplane and has occurred in a number of instances around major airports throughout Europe. Mosquitoes can survive a flight even in non-pressurized wheel bays . In a series of 29 respective patients from 1969 to 1988, most cases occurred within a distance of 2 km from the next airport, but longer distances are also possible when vectors are transported further, for example in a car . In this series, four patients died due to a delayed or missed diagnosis and approximately one third of the cases were persons working at or around the airport. A case of "port" malaria has been observed about 3.5 km off the port of Ghent, Belgium . In the secondary malaria patient, an "airport" malaria was considered due to a recent visit to Frankfurt airport, but dismissed after genotyping results revealed homology with the index patient's parasite strain.
The possibility of carrying infective mosquitoes within pieces of baggage to non-endemic areas or to areas far from international airports has been suspected in a number of previous cases [13, 20, 21]. The homology of strains in the two patients could accord with transmission by the same mosquito. In this scenario, the mosquito was carried in the baggage from Ivory Coast to Berlin, infected the index case before or during travel or at home in Berlin and then – after being transported to the hospital – infected the secondary patient.
Nosocomial P. falciparum infection, in which blood or parasite-containing fluids from a parasitaemic patient enter the bloodstream of a secondary patient, are well-documented. Example routes include needlestick injuries [22–26], incorrect use of blood glucose meters , or failure to exchange needles using multidose drug vials [28–30]. The incubation time following direct blood contact or needlestick injuries ranges from 4 to 17 days (median 12 days) .
There is also evidence of transmission through contaminated surfaces. This has been implicated in cases of malaria following re-use of catheters for contrast media  or handling of intravenous lines with contaminated gloves  and suggests that the infection can be passed by an intermediate and not necessarily visibly blood contaminated object. However, transmission in the absence of percutaneous or intravenous breaches has been described only in cases of direct contact with visible and infected blood. Published examples include a nurse with fingertip skin scratches (caused by peeling potatoes) , and a doctor with a minor nail-cutting injury; both became infected after having venosected malaria patients . In the broader nosocomial context, self-inoculation of patients' blood as a consequence of a psychiatric disorder has been reported from a German hospital, producing a series of thirteen malaria episodes in a laboratory technician .
The secondary case denied any needlestick injuries, invasive medical procedures performed on herself, or any contact between skin breaks and visible blood or contaminated fluids from the index case. However, her recall may not have been perfect, as she was interviewed nearly a month after the possible exposure time.
Probable route of transmission
In the present case of falciparum malaria acquired locally in Berlin, Germany, the route of transmission remains obscure despite extensive investigations. Transmission by local Anopheles on the hospital compound seems improbable in view of the timeline. While incubation times would allow for mosquito-borne transmission in the residential area, this option would require combinations of per se unlikely events. In addition, the absence of genetic recombination renders a mosquito-borne transmission unlikely. "Baggage malaria" cannot be ruled out but also appears unlikely: the mosquito would have had to survive for 11 days in the flat in Berlin (all bags had been emptied), re-enter the baggage in which it is was imported from Ivory Coast, to be carried to the hospital where it infected the secondary case. Nosocomial transmission would require fewer improbable events but suffers from the lack of a clear exposure history for the secondary case.
Implications for public health management
Because of the high fatality of undiagnosed falciparum malaria, suspected mosquito-borne transmission in non-endemic areas has significant public health implications. Protecting the population, even against a hypothetical threat, has highest priority under such circumstances.
In the case described here, mosquito-borne transmission was assumed in the early stages of investigation. It had to be considered that at least one infectious A. maculipennis s.l. could pose a danger during the rest of its lifespan (approximately six weeks, < 1 blood meal/week). In case of "baggage malaria", Anopheles gambiae would have been present for at least a few days on the hospital compound (life span approximately three weeks, one blood meal every other day). Careful information management and adequate risk communication in such a case is essential: missing potential further cases of autochthonous malaria must be avoided, but the extensive media coverage that might have followed a wider alert was also undesirable. In particular, with extensive media coverage, a high number of febrile patients with other diagnosis would need to undergo malaria testing. Here, the policy of selectively informing key health facilities and physicians who have initial contact with febrile patients proved to be adequate. In parallel, preparedness for potential subsequent events included surplus procurement of rapid test devices, allocation of further diagnostic capacity (staff, infrastructure), daily exchange of information between stakeholders, and training of personnel primarily involved in case of upward disease activity. Generally, in places outside endemic areas but with temporal presence of vectors competent for transmitting malaria, artemisinin-based anti-malarial therapy may contribute to reduce the risk of local transmission due to the gametocidal activity of artemisinins .
Key recommendations for public health management in malaria infections outside endemic areas
• Rapid and regular assessment of new information by a team of clinicians, malaria experts, entomologists and public health experts
• Timeline of events with incubation times for different transmission scenarios to assess their probability
• Molecular genotyping of isolates from index- and secondary cases
• Repeated interviews of index- and secondary cases as well as all other personnel involved
• Local entomological survey
• Selectively informing key health infrastructures and physicians
• Preparation of a press release in case of public media attention
In conclusion, the route of transmission in this unusual case of falciparum malaria in Berlin could not be determined. With climate change improving conditions for competent anopheline vectors in non-endemic areas, the risk of isolated P. falciparum infections or small outbreaks secondary to imported cases increases. This possibility should specifically be considered in large cities with high numbers of returning travellers and semi-immune migrants, who may carry infective gametocytes for extended periods of time. Due to its gametocidal activity, artemisinin-based anti-malarial therapy should be preferred in all settings with a competent vector population. Direct routes of transmission, such as nosocomial transmission or, rarely, self-inoculation should be considered in cases of otherwise unexpected P. falciparum infection. Genetic comparison of malaria species can be helpful in evaluating transmission scenarios. Even in non-endemic areas and in the absence of a travel history, it is important to consider malaria in patients with unexplained fever.
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.
- Centralized information system for infectious diseases (CISID) – WHO Regional Office for Europe. [http://data.euro.who.int/cisid]
- Dalitz MK: Autochthone Malaria im Mitteldeutschen Raum. PhD thesis. 2005, Martin-Luther-Universität Halle-Wittenberg, Medical Faculty
- Multifocal autochthonous transmission of malaria – Florida, 2003. MMWR Morb Mortal Wkly Rep. 2004, 53: 412-413.
- Snounou G, Viriyakosol S, Zhu XP, Jarra W, Pinheiro L, do Rosario VE, Thaithong S, Brown KN: High sensitivity of detection of human malaria parasites by the use of nested polymerase chain reaction. Mol Biochem Parasitol. 1993, 61: 315-320. 10.1016/0166-6851(93)90077-B.View ArticlePubMed
- Snounou G, Zhu X, Siripoon N, Jarra W, Thaithong S, Brown KN, Viriyakosol S: Biased distribution of msp1 and msp2 allelic variants in Plasmodium falciparum populations in Thailand. Trans R Soc Trop Med Hyg. 1999, 93: 369-374. 10.1016/S0035-9203(99)90120-7.View ArticlePubMed
- Su XZ, Carucci DJ, Wellems TE: Plasmodium falciparum: parasite typing by using a multicopy microsatellite marker, PfRRM. Exp Parasitol. 1998, 89: 262-265. 10.1006/expr.1998.4299.View ArticlePubMed
- Taylor HM, Kyes SA, Harris D, Kriek N, Newbold CI: A study of var gene transcription in vitro using universal var gene primers. Mol Biochem Parasitol. 2000, 105: 13-23. 10.1016/S0166-6851(99)00159-0.View ArticlePubMed
- Becker NPD, Zgomba M, Boase C, Dahl C, Lane J, Kaiser A: Mosquitoes and their Control. 2003, New York, Boston, Dordrecht, London, Moscow: Kluwer Academic/Plenum PublishersView Article
- Kotter H: Bionomie und Verbreitung der autochtonen Fiebermücke Anopheles plumbeus (Culicidae) und ihrer Vektorkompetenz für Plasmodium falciparum, Erreger der Malaria tropica. PhD thesis. 2005, University of Heidelberg, Faculty of Natural Sciences
- Knottnerus OS: Malaria Around the North Sea: A Survey. Climatic Development and History of the North Atlantic Realm: Hanse Conference Report. Edited by: Gerold Wefer WHB, Karl-Ernst Behre, Eynstein Jansen. 2002, Berlin, Heidelberg: Springer-Verlag, 339-353.View Article
- Martini E: Über die Malaria-Epidemie an der Nordseeküste 1826. Zeitschrift für Hygiene und Infektionskrankheiten. 1938, 12: 36-43.
- Kruger A, Rech A, Su XZ, Tannich E: Two cases of autochthonous Plasmodium falciparum malaria in Germany with evidence for local transmission by indigenous Anopheles plumbeus. Trop Med Int Health. 2001, 6: 983-985. 10.1046/j.1365-3156.2001.00816.x.View ArticlePubMed
- Mantel CF, Klose C, Scheurer S, Vogel R, Wesirow AL, Bienzle U: Plasmodium falciparum malaria acquired in Berlin, Germany. Lancet. 1995, 346: 320-321. 10.1016/S0140-6736(95)92212-1.View ArticlePubMed
- Do Rosario VE, Vaughan JA, Coleman RE: Analysis of the sporogonic development of Plasmodium falciparum and Plasmodium berghei in anopheline mosquitoes. Parassitologia. 1989, 31: 101-111.PubMed
- Mohrig W: Die Culiciden Deutschlands. Parasitologische Schriftreihe. 1969, 18: 260-
- Verhage DF, Telgt DS, Bousema JT, Hermsen CC, van Gemert GJ, Meer van der JW, Sauerwein RW: Clinical outcome of experimental human malaria induced by Plasmodium falciparum-infected mosquitoes. Neth J Med. 2005, 63: 52-58.PubMed
- Russell RC: Survival of insects in the wheel bays of a Boeing 747B aircraft on flights between tropical and temperate airports. Bull World Health Organ. 1987, 65: 659-662.PubMed CentralPubMed
- Isaacson M: Airport malaria: a review. Bull World Health Organ. 1989, 67: 737-743.PubMed CentralPubMed
- Peleman R, Benoit D, Goossens L, Bouttens F, Puydt HD, Vogelaers D, Colardyn F, Woude Van de K: Indigenous malaria in a suburb of Ghent, Belgium. J Travel Med. 2000, 7: 48-49.View ArticlePubMed
- Praetorius F, Altrock G, Blees N, Schuh N, Faulde M: Imported Anopheles: in the luggage or from the airplane? A case of severe autochthonous malaria tropica near an airport. Dtsch Med Wochenschr. 1999, 124: 998-1002. 10.1055/s-2007-1024464.View ArticlePubMed
- Rizzo F, Morandi N, Riccio G, Ghiazza G, Garavelli P: Unusual transmission of falciparum malaria in Italy. Lancet. 1989, 1: 555-556. 10.1016/S0140-6736(89)90095-0.View ArticlePubMed
- Herwaldt BL, Juranek DD: Laboratory-acquired malaria, leishmaniasis, trypanosomiasis, and toxoplasmosis. Am J Trop Med Hyg. 1993, 48: 313-323.PubMed
- Alweis RL, DiRosario K, Conidi G, Kain KC, Olans R, Tully JL: Serial nosocomial transmission of Plasmodium falciparum malaria from patient to nurse to patient. Infect Control Hosp Epidemiol. 2004, 25: 55-59. 10.1086/502293.View ArticlePubMed
- Needlestick malaria with tragic consequences. Commun Dis Rep CDR Wkly. 1997, 7 (28): 247-
- Carriere J, Datry A, Hilmarsdottir I, Danis M, Gentilini M: Transmission of Plasmodium falciparum following an accidental sting. Presse Med. 1993, 22: 1707-PubMed
- Haworth FL, Cook GC: Needlestick malaria. Lancet. 1995, 346: 1361-10.1016/S0140-6736(95)92371-3.View ArticlePubMed
- Moro ML, Romi R, Severini C, Casadio GP, Sarta G, Tampieri G, Scardovi A, Pozzetti C: Patient-to-patient transmission of nosocomial malaria in Italy. Infect Control Hosp Epidemiol. 2002, 23: 338-341. 10.1086/502062.View ArticlePubMed
- Evans RJ: Nosocomial malaria. Lancet. 1997, 349: 574-10.1016/S0140-6736(05)64339-5.View ArticlePubMed
- Al-Saigul AM, Fontaine RE, Haddad Q: Nosocomial malaria from contamination of a multidose heparin container with blood. Infect Control Hosp Epidemiol. 2000, 21: 329-330. 10.1086/501765.View ArticlePubMed
- Gonzalez L, Ochoa J, Franco L, Arroyave M, Restrepo E, Blair S, Maestre A: Nosocomial Plasmodium falciparum infections confirmed by molecular typing in Medellin, Colombia. Malar J. 2005, 4: 9-10.1186/1475-2875-4-9.PubMed CentralView ArticlePubMed
- Tarantola A, Rachline A, Konto C, Houze S, Sabah-Mondan C, Vrillon H, Bouvet E: Occupational Plasmodium falciparum malaria following accidental blood exposure: a case, published reports and considerations for post-exposure prophylaxis. Scand J Infect Dis. 2005, 37: 131-140. 10.1080/00365540410020839.PubMed
- Chen KT, Chen CJ, Chang PY, Morse DL: A nosocomial outbreak of malaria associated with contaminated catheters and contrast medium of a computed tomographic scanner. Infect Control Hosp Epidemiol. 1999, 20: 22-25. 10.1086/501557.View ArticlePubMed
- Piro S, Sammud M, Badi S, Al Ssabi L: Hospital-acquired malaria transmitted by contaminated gloves. J Hosp Infect. 2001, 47: 156-158. 10.1053/jhin.2000.0907.View ArticlePubMed
- Borsch G, Odendahl J, Sabin G, Ricken D: Malaria transmission from patient to nurse. Lancet. 1982, 2: 1212-10.1016/S0140-6736(82)91219-3.View ArticlePubMed
- Burne JC: Malaria by accidental inoculation. Lancet. 1970, 2: 936-10.1016/S0140-6736(70)92117-3.View ArticlePubMed
- Kun JF, Kremsner PG, Kretschmer H: Malaria acquired 13 times in two years in Germany. N Engl J Med. 1997, 337: 1636-10.1056/NEJM199711273372220.View ArticlePubMed
- Adjuik M, Babiker A, Garner P, Olliaro P, Taylor W, White N: Artesunate combinations for treatment of malaria: meta-analysis. Lancet. 2004, 363: 9-17. 10.1016/S0140-6736(03)15162-8.View ArticlePubMed
- Jelinek T, Ed: TropNetEurop Sentinel Surveillance Report:. 2007, March , Berlin. Falciparum malaria in 2006
- Hanscheid T, Melo-Cristino J, Grobusch MP, Pinto BG: Avoiding misdiagnosis of imported malaria: screening of emergency department samples with thrombocytopenia detects clinically unsuspected cases. J Travel Med. 2003, 10: 155-159.View ArticlePubMed
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.