SEARCH

SEARCH BY CITATION

Keywords:

  • Africa;
  • Americas;
  • cholera;
  • environmental microbiology;
  • epidemiology;
  • molecular epidemiology;
  • transmission

Abstract

  1. Top of page
  2. Abstract
  3. Cholera in Hispaniola
  4. Cholera in the Lake Chad Basin
  5. Cholera in the Great Lakes region and in East Africa
  6. Cholera in West Africa
  7. Discussion
  8. Acknowledgements
  9. Transparency Declaration
  10. References

Clin Microbiol Infect 2012; 18: 231–238

Abstract

The cholera burden has grown strikingly during the past 4 years, and has spread to countries previously spared by this disease. The current spread has proved especially violent, as illustrated by the recent deadly epidemics around the Lake Chad Basin, in East Africa, and in Haiti. This onset of severe cholera epidemics is part of the overall dynamic of the current seventh cholera pandemic, composed of successive epidemic waves. The current wave is attributable to new atypical El Tor strains, which spread from the Bay of Bengal to Papua in the east, Africa, and the Caribbean Sea in the west, and caused hundreds of thousands of cases and thousands of deaths during each of the last 4 years. The particular severity of the resulting epidemics is partially attributable to the specific characteristics of the atypical El Tor strain involved. Besides the abilty of El Tor to spread easily, this strain is associated with more severe clinical findings, because of elevated levels of toxin secretion resulting from a genetic content originating from classical strains. Conversely, recent studies of these deadly outbreaks raised hope by illustrating their relationship with human-borne dissemination rather than with the resurgence of environmental strains. As human-borne dissemination can be more easily targeted than ubiquitous environmental contamination, accurate and comprehensive epidemiological studies are essential to better understand the dynamics of the disease and to optimize future cholera responses.

For the fourth consecutive year, cholera incidence increased in 2010. The WHO reported 317 534 cases worldwide in 2010, as compared with 177 963 in 2007 [1,2]. In the same period, the number of cholera-related deaths almost doubled (7543 vs. 4031). The year 2011 is expected to be even worse, as large epidemics are ongoing in several countries, including Haiti, Nigeria, Cameroon, Chad, and the Democratic Republic of Congo (DRC). As a consequence, the World Health Assembly recognized this year the re-emergence of cholera as a significant global health problem, and adopted resolution WHA 64.1 calling for implementation of ‘an integrated and comprehensive global approach to cholera control’ [1]. A prerequisite for such control programmes is to precisely understand what is actually occurring in the main areas currently facing cholera epidemics.

In 2010, almost 90% of the total number of cholera cases reported worldwide were notified in three foci (Fig. 1): the two countries in Hispaniola (Haiti and the Dominican Republic), the four countries of the Lake Chad Basin in Africa (Nigeria, Cameroon, Niger, and Chad), and seven countries in either the Great Lakes region or on the eastern coast of Africa (east of the DRC, Uganda, Kenya, Burundi, the United Republic of Tanzania, Zambia, and Mozambique). In contrast, countries known to suffer from endemo-epidemic cholera (e.g. in West Africa from Mauritania to Sierra Leone, including Gambia, Guinea Bissau, and Guinea Conakry) experienced in 2010 an almost total collapse of cholera transmission. Presented here is an overview of the dynamic of the cholera epidemic in each of these countries, focusing on determining the importance of environmental and human-borne mechanisms of transmission.

image

Figure 1.  World repartition of cholera cases, 2010. Details on the three major foci: (a) Hispaniolia; (b) Lake Chad Basin; (c) Great Lakes and eastern coast of Africa.

Download figure to PowerPoint

Cholera in Hispaniola

  1. Top of page
  2. Abstract
  3. Cholera in Hispaniola
  4. Cholera in the Lake Chad Basin
  5. Cholera in the Great Lakes region and in East Africa
  6. Cholera in West Africa
  7. Discussion
  8. Acknowledgements
  9. Transparency Declaration
  10. References

Cholera appeared in Haiti in October 2010, probably for the first time in its history [3]. One year later, the cholera epidemic in Haiti has become the largest epidemic of cholera in the last 2 decades, with 476 714 recorded cases and 6648 deaths as on 18 October 2011, just 1 year after the epidemic started. The current situation in Haiti is even worse than in Peru during the 1991 cholera epidemic (322 562 cases and 2909 deaths at the end of 1991, after 11 months of the epidemic) [4]. Despite its better infrastructure, the neighbouring Dominican Republic is experiencing a spill-over of the Haiti cholera epidemic, with 17 758 suspected cases and 308 deaths recorded between October 2010 and September 2011 [5].

Although investigating such a deadly cholera epidemic was deemed to be of crucial interest [6], only one peer-reviewed publication has described a field investigation detailing the initial spread of cholera throughout Haiti [7]. This investigation showed an exact correlation in time and place between the arrival of a peacekeeper battalion during a UN troop rotation and the onset of the epidemic in a hamlet, called Meye, located in a remote area of Centre Department, Haiti [7]. The peacekeeping soldiers stationed in Meye came from Nepal, which was experiencing a cholera outbreak at that time [1]. According to a panel of scientists appointed by the UN, ‘the source of the Haiti cholera outbreak was due to contamination of the Meye Tributary of the Artibonite River with a pathogenic strain of current South Asian type Vibrio cholerae as a result of human activity. This contamination initiated an explosive cholera outbreak downstream in the Artibonite River Delta, and eventually, throughout Haiti’ [8]. The deadly epidemic in the Artibonite River Delta provoked a panic that made many people flee to neighbouring communities. Soon after, these areas were experiencing cholera outbreaks and, within 2 months, almost all Haiti was rife with cholera (Fig. 2). Several tools were used to characterize the V. cholerae isolates collected during the Haitian cholera epidemic, comparing them with isolates from other epidemics. These tools were pulsed-field electrophoresis, variable-number tandem-repeat analysis (also called multiple-locus variable-number tandem-repeat analysis analysis), or partial genome sequencing, or whole genome sequencing [9–16]. The clonal aspect of the isolates collected in Haiti and their low level of diversification argue against an environmental origin of the Haitian cholera epidemic [9]. Even though, prior to 2010, V. cholerae had never been isolated from Haitian estuaries or river waters, environmental strains were initially incriminated in the Haitian epidemic, as V. cholerae was known to be a natural inhabitant of aquatic environments [17,18]. Culturable and viable but non-culturable V. cholerae were found attached to aquatic zooplankton and phytoplankton, especially in estuarine areas [19]. In the Bay of Bengal, cholera incidence was influenced by local factors, including rainfall, variation in river levels, and blooms of plankton, and by more global climatic conditions, such as the increase in sea surface temperatures linked to El Niño events [20–22]. None of these conditions was found to be associated with the onset of the cholera epidemic in Haiti.

image

Figure 2.  Expansion of cholera, Haiti, fourth trimester 2010. The whole country was hit in 2 months.

Download figure to PowerPoint

Besides their clonal origin and the recent findings that Haitian cholera isolates were identical to Nepalese isolates collected during the same period [14], Haitian V. cholerae specimens presented particular characteristics that might partially explain the severity of the epidemic in Haiti. According to Cecarelli et al. [11], specimens collected in Haiti contained an integrative conjugative element of the SXT/R391 family, a major drug resistance-spreading vector in bacteria. By sequencing the gene coding for the cholera toxin subunit B, these authors also showed that the Haitian isolates were harbouring a genotype that was previously found in an altered El Tor V. cholerae variant isolated during the harsh cholera epidemic in Orissa, India, in 2007 [11]. Interestingly, a very close variant of this strain was incriminated in another severe epidemic in 2010. This epidemic involved several countries around the Lake Chad Basin and, according to the WHO, was responsible for 62 762 cases, including 2610 deaths, or 54% of cases and 77% of deaths recorded in Africa by the WHO in 2010 [1].

Cholera in the Lake Chad Basin

  1. Top of page
  2. Abstract
  3. Cholera in Hispaniola
  4. Cholera in the Lake Chad Basin
  5. Cholera in the Great Lakes region and in East Africa
  6. Cholera in West Africa
  7. Discussion
  8. Acknowledgements
  9. Transparency Declaration
  10. References

During 2010, the cholera situation in Africa was highly variable. On the one hand, a deadly epidemic affected the Lake Chad Basin, and several outbreaks occurred in the Great Lakes region and on the East African coast. On the other hand, the remainder of Africa experienced a relatively quiescent period (which, however, sometimes proved transitory, according to 2011 reports).

The cholera epidemic around Lake Chad started in April 2010, and faded by the end of the year. Very few cases were reported in Cameroon and Nigeria from January to early April 2010, and no cases were notified in Chad and Niger [23]. The first months of 2010 were thus considered to be an inter-epidemic period, following the 2009 epidemic, which mainly involved the states of Adamawa and Borno in Nigeria and the northern region of Cameroon. The 2010 epidemic started with a weekly incidence in early July in both Cameroon and Nigeria foci of approximately 200 cases. This period was also marked by a spread of cholera, reaching Chad and Niger in July. The situation worsened in July, as the weekly incidence of cholera increased simultaneously in the four countries, reaching a peak in August or September, depending on the affected nation. The worst 7 days occurred during the 37th epidemic week (13–19 September), with almost 7000 new cases recorded altogether in the four countries. Thereafter, the epidemic extended beyond national borders, with the epidemic macro-centre being by the lakeside districts of Lake Chad [23] (Fig. 3). Thereafter, the number of new cases dropped rapidly, to reach a level similar to that in the pre-epidemic period at the end of 2010, except in Cameroon, where more than 100 cases a week were still recorded in December.

image

Figure 3.  Evolution of cholera, Lake Chad Basin, 2010, based on the map published in Oger and Sudre, 2011 [23]. Four countries were concerned (Chad, Niger, Nigeria, and Cameroon) from 1 January 2010 to 17 October 2010. Period 1: from 1 January (week 1) to 4 April (week 14). Inter-epidemic period: a few cases only persist in three local government areas on the lake shores (Nigeria). Period 2: from 5 April to 4 July (week 27). Beginning of the epidemics: four major foci experienced an increase of cases, one on the west of Lake Chad (Niger and Nigeria), one in the Mayo-Sava and Mayo-Tsanaga departments (Cameroon), another in the Mayo-Danay department (Cameroon), and the last one around Adamawa (Nigeria). Period 3: from 5 July to 22 August (week 33). Ascension of the epidemic curve, without new major foci. Period 4: from 23 August to 17 October (week 41). Epidemic peak and maximal geographical extension of the 2010 epidemic.

Download figure to PowerPoint

Very little is known about what triggered this regional epidemic. Indeed, while the epidemic was occurring, the other Central and West African countries were experiencing a relative lull, with limited cholera outbreaks and a marked decrease in the number of deaths. The Lake Chad epidemic appeared to be correlated with the occurrence of the rainy season in the four concerned countries. The epidemic started earlier and continued for longer in northern Nigeria and Cameroon, where the rainy season lasts from late June to September, than in southern Chad and Niger, where the rainy season is shorter. However, nothing in the UNICEF report suggested specific climatic changes or upheavals in this region during 2010.

The importation of a more toxigenic new strain could be an explanation for this particularly severe African epidemic. In a letter published in 2011, Quilici et al. reported that the 2009 cholera epidemic in Nigeria and Cameroon involved an atypical El Tor strain with a classical cholera toxin gene, similar to the strain that appeared in Orissa, India, and that has gradually replaced the prototype El Tor strain in South Asia since 2007 [24]. The importance of strain dissemination in the dynamics of worldwide cholera epidemiology was additionally highlighted by more recent studies showing that the cholera strain involved in the deadly Haitian epidemic was also an atypical El Tor strain, similar to strains from recent Cameroon or Indian epidemics [15].

Cholera in the Great Lakes region and in East Africa

  1. Top of page
  2. Abstract
  3. Cholera in Hispaniola
  4. Cholera in the Lake Chad Basin
  5. Cholera in the Great Lakes region and in East Africa
  6. Cholera in West Africa
  7. Discussion
  8. Acknowledgements
  9. Transparency Declaration
  10. References

According to the WHO, in 2010 the number of reported cases declined in the Great Lakes region and on the East African coast as compared with 2009. Nevertheless, these regions remain a major focus of cholera, with 13 384 cases reported in eastern DRC, 7430 in Mozambique, 6794 in Zambia, 4469 in Tanzania, 3188 in Kenya, 2341 in Uganda, and 333 in Burundi. No cases were reported in Rwanda. Taken altogether, these countries reported 37 939 cases and 399 deaths in 2010. Recent studies have shown that lakeside areas located in the Great Lakes region play a role in the regional persistence of cholera, being the source of cholera outbreaks that can potentially spread into neighbouring territories [25–27]. The mechanisms underlying cholera resurgences in these lake areas remain unclear. Some have proposed a relationship with plankton blooms [28] or water hyacinths [29], similar to the recognized pattern of cholera in the Bay of Bengal. However, the increase in cholera incidence in the rainy season might also be related to changes in human exposure through sanitation and food consumption. Sewage seepage may have resulted in more frequent contamination of water resources, and higher fish consumption and seasonal changes in human activities, as in the east of the DRC [27], may have contributed to increased exposure to bacteria. Such exposures, rather than a natural rise in aquatic the environmental Vibrio concentration, may have led to local and regional outbreaks. Recently, epidemics appeared in areas in which the population had previously been spared, e.g. in Kinshasa and the western provinces of the DRC. This spread, which followed the Congo River, was probably related to human displacements, as the river is the main source of human communication and interaction. Thus, more than 1000 cholera cases were reported in 2011 from the Equateur region in the north of the DRC, a territory where cholera had rarely been documented (personal information). Whatever its mechanisms, this spread is a worrying trend that allowed the disease to reach Kinshasa, the country’s highly populated capital.

Cholera in West Africa

  1. Top of page
  2. Abstract
  3. Cholera in Hispaniola
  4. Cholera in the Lake Chad Basin
  5. Cholera in the Great Lakes region and in East Africa
  6. Cholera in West Africa
  7. Discussion
  8. Acknowledgements
  9. Transparency Declaration
  10. References

Contrasting with the spread of cholera into previously spared areas, some countries were experiencing prolonged quiescent periods. Such a phenomenon is currently being observed in the western extremity of Africa, including Senegal, Gambia, Guinea Conakry, and Guinea Bissau. Until 2009, these countries were regularly impacted by cholera outbreaks, which started in coastal areas of Guinea Conakry and Guinea Bissau during the rainy season (e.g. from May to November in Guinea Conakry). The iterative onset of cholera during the warm and rainy season was sometimes considered to be a strong argument linking cholera with a permanent environmental reservoir constituted by lagoons and estuaries. However, in large cities such as Conakry, cholera transmission during the 2004–2007 rainy seasons may alternatively have been facilitated by the contamination of water distributed through a supply network of porous pipes. The current quiescent period, with almost no case of cholera for three consecutive years, suggests that the Guinean coastal environment is not a permanent reservoir that yearly triggers seasonal cholera outbreaks. Interestingly, the almost total disappearance of cholera in Guinea Conakry and Guinea Bissau is contemporaneous with an improvement in the cholera situation in Sierra Leone and Liberia, two countries located east of Guinea Conakry that experienced large cholera outbreaks from 2003 to 2007. To distinguish between local environmental resurgences of varied cholera strains and iterative re-introduction by human activity of monoclonal or oligoclonal cholera strains, studies focusing on strain genotyping must be performed [9].

On the opposite side of inter-tropical Africa, a similar phenomenon occurred in Kenya. There, the number of cholera cases also dropped dramatically after the end of the last large epidemic (14 613 suspected cases and 327 deaths in 2009 and during the first months of 2010), with only a few imported cases detected in refugee camps in the north of the country. However, cholera was recently brought back in the refugee camps in northern Kenya, with 60 confirmed cases on 15 November. This re-introduction of cholera was apparently related to the arrival of new infected refugees [30], which highlighted once again the importance of human displacements in the ongoing spread of the disease.

Discussion

  1. Top of page
  2. Abstract
  3. Cholera in Hispaniola
  4. Cholera in the Lake Chad Basin
  5. Cholera in the Great Lakes region and in East Africa
  6. Cholera in West Africa
  7. Discussion
  8. Acknowledgements
  9. Transparency Declaration
  10. References

Cholera is still an epidemic disease whose spread into vulnerable territories is responsible for severe outbreaks. Numerous examples highlight this current phenomenon, notably the deadly epidemic in Haiti, the spread of cholera in the Lake Chad Basin from an initially restricted focus in Nigeria and Cameroon, and the travel of cholera from the east of the DRC to the suburbs of Kinshasa 1500 km away, following the Congo River.

The environment plays an important role in cholera transmission, even far from the Bay of Bengal, as shown by the studies in coastal and lake areas in Africa. However, the role of aquatic environments as a permanent reservoir serving as the source of iterative epidemics is still debatable. The difficulties in predicting the evolution of cholera according to local environmental characteristics is illustrated by the Americas. In 2000, an ecological model suggested a high risk of endemic cholera in coastal areas of Mexico [31]. However, Mexico has remained free from cholera for a decade [1]. Overall, despite the epidemic waves of the 1990s, Latin America as a whole is currently almost free of cholera, with the dramatic exception of human-induced cholera in Hispaniola [1]. Residual environmental toxigenic V. cholerae strains probably exist here and there, and cause occasional human contamination, but no significant epidemic has been reported for over a decade.

Madagascar is, in many respects, similar to Hispaniola. It is an island with deficient sanitation, a susceptible hydrogeological environment, widespread water-fed rice paddies, political tensions, and lack of resources, that was plagued by successive cholera waves from 1999 to 2001. However, Madagascar has not experienced any new outbreak in the ensuing 10 years. This evolution from highly epidemic cholera to absent cholera, previously described in West Africa, raise hopes that the disease might be eliminated or at least greatly reduced, and undermine the fatalistic hypothesis that countries experiencing severe epidemics are doomed to live forever with the pathogen, owing to permanent environmental contamination. The fact that the recent severest cholera outbreaks originated in inland territories such as central Haiti and the African lakes region suggests that the epidemiological link between clinical cholera and estuarine environments is weak outside its initial source in the Bay of Bengal. The importance of rivers and fresh water for environmental suitability for the cholera organism is the subject of contemporary study and debate, as cholera epidemiology is proving to be more complicated than just the product of a relationship between sea surface temperatures of estuarine waters and bacterial resurgence [32].

The role of new variants in cholera epidemics should be noted. The three most severe cholera epidemics in the last 4 years were caused by a new, atypical El Tor strain, which had a potentially more toxigenic genetic content than usually found in classical strains. The so-called seventh cholera pandemic, which was alleged to have started in Indonesia in 1961, was in fact composed of several waves of cholera transmission arising from the Bay of Bengal [33]. The new atypical variant responsible for the most recent wave proved to be able to gradually replace the classical El Tor strain in South Asia [34–36] and to spread into Central Africa and Caribbean Sea in <4 years. This atypical El Tor strain was also involved in the Zimbabwean epidemic in 2009 [37]. Eastwards, an altered El Tor strain also reached Papua New Guinea from Vietnam in 2009, and caused more than 15 000 cases [38]. This strain replacement is worrying, as the atypical strain found from India to Haiti is associated with increased levels of cholera toxin production [16,39], less frequent asymptomatic infection, and more severe outcome of the disease [40].

Given the changing nature of contemporary cholera strains, there is a crucial need to better monitor the evolution of cholera all around the world. Indeed, cholera epidemiology faces a new paradigm shift. The first change of paradigm took place in the 1960s, when the perception of cholera shifted from a disease caused by interpersonal transmission to one that had strong relationships with environment and natural phenomena. This perception led many to assume that cholera is an almost unavoidable consequence of poverty in susceptible settings. However, recent studies on the spread of cholera have highlighted the actual importance of human-borne dissemination. The importance of the early identification of potential cholera carriers has become re-apparent as a key issue in preventing cholera spread, as was notably illustrated in Haiti [7,8]. This point has also been regularly stressed in WHO reports [1].

Unfortunately, national reporting systems for cholera are often woefully inadequate. The International Health Regulations followed by the WHO no longer require notification of cholera cases. Thus, the numbers of cholera cases reported to the WHO by various resource-constrained countries are highly underestimated. Many countries are, indeed, reluctant to notify cases, and choose inadequate reporting strategies. Some, such as the Dominican Republic, report only confirmed cases, whereas others, such as Bangladesh, reported no cholera cases in either 2009 or 2010, although it is a well-known endemic focus of cholera [1,41]. Therefore, the real worldwide burden of cholera is still unknown, and will not be known until the notification of suspected and confirmed cases becomes mandatory. Besides better overall numbers, more precise geographical data are needed. Indeed, mapping cases of cholera and analysing how the locations of cases are evolving in space and time is of crucial importance to understand the dynamics of cholera epidemics. It is important to remember that fieldwork and mapping of cholera cases in the Soho neighbourhood of London helped Dr John Snow to demonstrate the role of the Broad Street pump as the source of cholera in the area [42]. For his persistent efforts to determine how cholera is spread and for the mapping methods that he employed, John Snow is widely considered to be the father of epidemiology. What John Snow could do with active footwork and a basic map and pen, contemporary epidemiologists should be able to do with detailed field investigations, current Geographical Information Systems, recent spatial and geographical statistical developments, and phylogenic analysis of cholera strains through pulsed-field electrophoresis and genotyping. Unfortunately, despite the major public health challenge represented by cholera, analysis of the recent medical literature suggests that the use of modern analytical tools is still rare, notably in Africa. Given the thousands of lives that are at stake, this must change.

References

  1. Top of page
  2. Abstract
  3. Cholera in Hispaniola
  4. Cholera in the Lake Chad Basin
  5. Cholera in the Great Lakes region and in East Africa
  6. Cholera in West Africa
  7. Discussion
  8. Acknowledgements
  9. Transparency Declaration
  10. References