The AIDS epidemic begins, 1900-60
Available evidence provides a pretty good picture of where and when the AIDS epidemic began. How and why it began when it did are questions for which there are so far no agreed answers but only competing hypotheses. These hypotheses deal not only with history, but are also linked to ideas about what currently drives HIV epidemics in Africa, Asia, and the Caribbean. Before discussing hypotheses for the origin of the HIV epidemic, it is useful to review what is known about HIV’s early history.
Clues to the early history of the AIDS epidemic
The AIDS epidemic began decades before doctors first recognized AIDS in 1981. In 1983, French scientists discovered HIV, the virus that causes AIDS. Tests based on HIV soon found similar viruses in African monkeys, and later in apes (chimpanzees and gorillas). These viruses have been named simian immunodeficiency viruses (SIVs). Monkeys and chimpanzees infected with SIVs in the wild seldom if ever get sick, which is common when a virus has lived with a host species for many years.
Tests based on HIV also discovered a closely related virus circulating in humans. This virus was named HIV type 2 (HIV-2), while the first HIV discovered was renamed HIV-1. HIV-2 transmits much less efficiently than HIV-1 (through blood, sex, and mother-to-child), has not spread much outside West Africa, and leads more slowly and less predictably to AIDS. Although HIV-2 has much less impact on human health than does HIV-1, the origin and early history of HIV-2 provide insights into the parallel origin and early history of HIV-1.
Much of what we know about the pre-1981 history of the AIDS epidemic comes from analyses of HIV and SIV molecules. Like other viruses, each HIV and SIV is a large molecule composed of a sequence of smaller molecules (nucleotides). For various reasons – including random errors during virus reproduction – these sequences change over time. Comparing one virus with others, those whose sequences are the most alike are the fewest years removed from a common ‘parent’ virus which lived and multiplied in an infected host years ago. Moreover, estimates of the rate of change allow scientists to estimate the date of the most recent (last) common ancestor for any two HIVs and/or SIVs. From this information, they can draw viral ‘family trees,’ and can date the nodes which identify the last common ancestors of any two or more viruses.
HIV-1 origin and epidemic expansion to 1960
Where and when HIV-1 began
All HIV-1 molecules are more similar to – and thus more closely related to – SIV from chimpanzees and gorillas than to HIV-2 or to SIVs from other simian species. Hence, HIV-1 appears to have descended from SIV that passed from chimpanzees and gorillas to infect humans. (Once an SIV infects a human, it is an HIV. The change in name refers to the host. Whether SIV must change or evolve to transmit among humans is a matter of debate.)
All known HIV-1s can be sorted into three groups. The M (main) group accounts for well over 99 per cent of HIV-1 in the world. The O (outlier) group has been part of the HIV-1 epidemic in Cameroon and Gabon, but is rare in other countries. HIV-1 in the N (non-M, non-O) group has been found in a handful of people in Cameroon. Each of these three groups appears to have begun from a separate event in which SIV passed from a chimpanzee (twice) and a gorilla (once) to infect a human. The evidence for this is that HIV from each group is more similar to – and hence more closely related to – some of the SIV samples collected from chimpanzees and/or gorillas than to HIV from the other two groups.
All the SIV samples that are closest to HIV-1 in the M (main) and N groups have been collected from the chimpanzee sub-species, Pan troglodytes troglodytes, which ranges south of the Sanaga River in Cameroon, through Gabon and the Republic of Congo (hereafter identified as Congo) and east as far as the Congo river and its tributaries that define the northwest border of the Democratic Republic of Congo (DRC) (see the map of Africa in the Statistical Annex). More distantly related SIV has been found in the chimpanzee sub-species, P.t. schweinfurthii, which ranges east of the Congo River across DRC into Tanzania. Because chimpanzees are loath to cross rivers, these two sub-species, living on opposite sides of the Congo River, have bred separately for an estimated 117,000 years.[i] SIV was circulating among chimpanzees before these two populations split, and has therefore been available to infect humans for well over 100,000 years.
In a remarkable bit of scientific sleuthing, scientists found close matches between SIV from the feces of wild chimpanzees living in southeast Cameroon and the HIV-1 M group.[ii] Apparently, the virus that accounts for most of the world’s HIV infections passed from a chimpanzee to a human somewhere in southeast Cameroon. The same study found other close matches between SIV from chimpanzees in south-central Cameroon and the HIV-1 N group.
In 2006, researchers reported SIV in gorilla feces collected in southern Cameroon.[iii] From analyses of SIV sequences, the gorilla’s SIV appears to be a branch of the viral tree for SIV from chimpanzees, which suggests that a gorilla many centuries ago somehow contracted SIV from a chimpanzee. Surprisingly, HIV-1 from the O group appears to be most closely related to SIV from gorillas. Apparently SIV from a gorilla passed to a human somewhere in or near southern Cameroon to begin the HIV-1 O group. So far none of the SIV sequences from gorillas are close enough to O group sequences to allow one to be more specific about where this might have occurred.
From differences between HIV-1 M group viruses, scientists have estimated the date of the most recent common ancestor of the M group. Ten estimates range over the period from 1902 to 1937.[iv] That is not the date when the SIV that begat the M group crossed to a human, but rather the date that an infected human – who may have been the first one infected, or someone later down the chain of infection – passed HIV to another to establish at least two lineages (chains of infection) that have survived and contribute to the current HIV-1 M group epidemic.
HIV-1 O group viruses are somewhat more diverse than M group viruses,[v] which suggests they have an older common ancestor. One estimated date for the most recent common ancestor of the O group is 1920.[vi] The N group shows much less diversity than the M or O groups, and therefore appears to have descended from a more recent parent virus,[vii] possibly in the 1950s.
Epidemic expansion to 1960
The Congo River and its tributaries link Cameroon with DRC and the Congo. Very likely someone with an early HIV-1 M group infection in Cameroon traveled downriver to DRC (which was at that time the Belgian Congo; this book uses place names in 2009, with occasional mention of historic names). HIV reached DRC very soon after the most recent common ancestor of the M group. We know this because the diversity in genetic sequences among M group viruses collected in DRC is comparable to the diversity found among all M group viruses from all over the world.
The earliest known HIV was found in one of approximately 700 blood samples collected from adults and children in various locations throughout DRC in 1959. The sample with HIV came from an adult male in Kinshasa.[viii] In 2008, scientists reported the discovery of another early HIV in lymph tissue collected from a woman in Kinshasa in 1960, and subsequently preserved in paraffin for almost five decades.[ix] Hundreds of kilometers away, in northern DRC, a woman was hospitalized in 1958 with ‘swollen lymph nodes, breathing difficulties, and dental problems,’[x] and died in 1962 with what was later identified (from symptoms, without a confirming test for HIV infection) to be AIDS.[xi] The earliest known O group infection was identified in a Norwegian sailor who visited Douala port in Cameroon in 1961-62. The sailor along with his wife and a child died from AIDS in 1976. Subsequent tests of stored tissues identified HIV-1 from the O group.[xii]
The fact that we can identify four unlinked HIV-1 infections or AIDS cases from Africa around 1960 suggests that HIV infections were not rare. By the 1960s, the HIV-1 M and O groups had spread to many people. How many?
Intriguingly, doctors in Central Africa reported normally rare infections during 1930-60 which were later found to be common in people with AIDS. For example, Thijs reported more than 200 cases of Kaposi’s sarcoma, a cancer caused by a virus, during 1939-55 in the DRC, along with other cases from Rwanda and Burundi.[xiii] In a substantial minority of these cases, Kaposi’s sarcoma appeared in people aged less than 30 years, and infected lymph nodes and internal organs, a presentation that has later been associated with AIDS. (In people without AIDS, Kaposi’s sarcoma is rare, and classically presents as infected patches of skin on the feet and legs of older men.) Doctors also reported cryptococcal meningitis and histoplasmosis, fungal infections later found to be common among people with AIDS.[xiv] However, there is no consensus about whether any of these pre-1960 infections in Africa point to early AIDS cases.
Molecular clues to early epidemic expansion
Viral sequences provide clues to early epidemic expansion. Most HIV-1 M group viruses can be sorted into one of nine subtypes or clades, which are identified by letters A through K (except E and I). The viruses in each of these clades are more closely related to each other – that is, their sequences are more similar – than to other HIV-1 M group viruses. The clades, in other words, are major limbs on the viral tree.
When one looks closely at HIV-1 sequences, many viruses are a mix of HIV-1 from two or more clades. These mixed viruses, or ‘recombinants,’ develop in people who have somehow acquired two or more HIV infections. Several dozen of these mixed viruses have spread to many people, establishing major branches on the viral tree. These common recombinants are called circulating recombinant forms, or CRFs.
Most of the founder viruses for clades and CRFs can be dated to the 1950s or later. Importantly, many HIV-1 M group viruses collected in Central Africa have no close relative among other viruses that have been sequenced – they cannot be assigned to any clade or CRF. In other words, each of these unclassified viruses identifies a separate chain of infection (branch of the viral tree) going back to the early decades of the M group.
The Los Alamos National Laboratory in the US collects and posts sequences of HIV-1 molecules from studies throughout the world in the HIV Sequence Database.[xv] As of October 2006, this Database listed over 100 unclassified HIV-1 M group samples from six countries in Central Africa. About 5 percent of samples were unclassified in DRC, 2 percent in Cameroon, and 3 percent in four other countries of the region (Central African Republic [CAR], Congo, Equatorial Guinea, and Gabon). Studies to date have sampled and sequenced HIV-1 from far less than 1 percent of Central Africans living with HIV-1 infections. If we could sequence HIV from all current infections, we would very likely find more than a thousand unclassified viruses descending from separate early branches (infections) in the first several decades of the epidemic.
There is also some evidence for early expansion in the HIV-1 O group. Relative to the M group, a larger proportion of O group sequences have no recent relatives. The O group viral tree, in other words, has many long branches – chains of infection – beginning in the first several decades after the O group began to spread among humans. Current low numbers of HIV-1 O group infections – possibly 10,000 in the world – are evidence for long periods of slow growth,[xvi] no growth, or even declines in the number of infections. Analyses of available sequences cannot distinguish between these different possibilities, and so provide little information about the numbers of O group infections during 1920-60.
HIV-2 origin and epidemic expansion to 1960
Just as HIV-1 comes from chimpanzees and gorillas, HIV-2 comes from SIVs that passed from sooty mangabeys – which are native to West Africa – to humans. As of 2009, all known HIV-2s can be sorted into eight groups, labeled A through H. Each group appears to have begun with a separate transmission of SIV from sooty mangabeys to humans. Groups A and B account for most infections. Group F has been found in two people,[xvii] while each of the other five groups, C, D, E, G, and H, has been found in only one infection.
Molecular analyses have found that HIV-2 from the A and B groups is most closely related to SIV from sooty mangabeys in southwestern Cote d’Ivoire, suggesting that these groups began with SIV transmissions to humans in Cote d’Ivoire or eastern Liberia.[xviii] As of 2009, the one available estimate for the date of the most recent common ancestor of the HIV-2 A group is 1940. However, this is based on only 17 samples, of which at least 13 came from Guinea-Bissau (a Portuguese colony till 1974) or Portugal. Similarly, the one available estimate for the date of the last parent of all HIV-2 B group viruses – 1945 – is even more weakly based on only five samples.[xix] Including more samples in these analyses will likely increase the range of sequence diversity, and thereby push the estimated dates of their most recent common ancestors back into the 1930s or earlier.
There is very little information to describe and date the spread of the HIV-2 A and B groups from Cote d’Ivoire and/or Liberia to other countries in West Africa. Rare cases of HIV-2 infection have been identified from stored blood samples collected in Cote d’Ivoire, Gabon, Mali, Nigeria, and Senegal in the period 1965-74.[xx] The best source of information may be from sequencing, but so far not enough HIV-2s have been sequenced to get a picture of early epidemic spread.
Because people can live with HIV-2 infection for decades, even surviving to old age, surveys in the 1990s and even later provide information on infections before 1960. Among countries, Guinea-Bissau has the highest HIV-2 prevalence and – based on infections in older persons[xxi] – may well have had the highest prevalence from before 1960.
Why did the HIV epidemic emerge in the 20th century?
As of 2009, sequencing of SIV and HIV has found 11 groups of HIV that descend from separate events in which SIV passed to a human – twice from chimpanzees, once from a gorilla, and eight times from sooty mangabeys. Recent studies in Central Africa report near matches between other bloodborne viruses (for example, simian foamy virus)[xxii] in humans and in simians, pointing to other recent simian-to-human transmissions. This evidence suggests that common events, such as a hunter, butcher, or cook getting simian blood into a cut, pass bloodborne viruses from simians to humans.
Because SIV passed from simians to humans at least 11 times in the recent past, this no doubt occurred in the distant past as well. SIV from chimpanzees – which started the HIV-1 M group – has been the most deadly for humans. Very likely SIV has passed from chimpanzees to humans hundreds of times in the more than 100,000 years during which humans lived near SIV-infected chimpanzees in Africa. In past centuries, no doubt some men and women with rare HIV infections from gorillas, chimpanzees, and sooty mangabeys sometimes transmitted HIV to others through wounds, scarification and other traditional skin-piercing practices, sex, or mother-to-child. But because the parent viruses for all HIV-1 and HIV-2 groups can be dated to the 20th century, we can conclude that – on average – each person infected with HIV before the 20th century infected less than one other person, so that rare infections did not multiply, but rather died out over time.
Hence, the puzzle that must be solved to explain the origin of the AIDS epidemic is not how SIV passed to humans, but what happened next, so that HIV passed from one human to another fast enough to survive and to spread among humans. This insight undermines explanations for the origins of the epidemic that focus on the particular time and route for SIV to pass from chimpanzees to humans.
Specifically, this insight undermines the hypothesis – articulated by Edward Hooper in The River[xxiii] – that feeding of SIV-contaminated oral polio vaccine to 900,000 children and adults in Central Africa during 1957-60 caused the HIV epidemic. Even if contaminated vaccine had infected one or more persons, that alone would not explain the epidemic, because SIV had likely reached and infected humans many times in past millennia without starting an epidemic. There are other problems with the hypothesis. There is no solid evidence that the vaccine was prepared in chimpanzee kidney cells. There is no evidence it was contaminated with SIV. Even if the vaccine had been contaminated, HIV transmits poorly through oral exposures. And the hypothesized passage of SIV to humans through polio vaccine during 1957-60 occurred years after the estimated dates of the most recent common ancestors of the HIV-1 M and O groups.
Three competing hypotheses attribute HIV’s more successful human-to-human transmission in the period 1900-60 than in previous centuries to: more risky sex, more blood exposures, and viral evolution. More than one of these hypotheses could be at least partially true, and other factors – including so far unsuspected factors – may have also contributed to the early and unnoticed expansion of the HIV epidemic.
Hypothesis 1: More risky sex accelerated HIV transmission
This seems to be the most popular hypothesis, in line with the widespread assumption that sexual transmission drives Africa’s HIV epidemics. Did Africans have more sexual partners during 1900-60 than in previous centuries? If so, did this accelerate HIV transmission enough to create an epidemic?
As John Seale, an expert on sexually transmitted diseases, pointed out in 1987, ‘Promiscuous sexual intercourse is not new in Africa, or in any other continent…’[xxiv] Even so, there is no evidence that HIV survived before 1900 among Africans with multiple sexual partners, such as polygamous chiefs, warriors, long-distance traders, or women in sex work. On the other hand, colonial governments conscripting men for labor and military duty disrupted African families during the early 20th century. But similar large disruptions had occurred in previous centuries, such as during the slave trade before 1850.
Did sexual behavior change accelerate HIV transmission to allow the HIV epidemic? The hypothesis is speculative at best. However, considering the limited information about sexual behavior in the decades and centuries before 1960, anything is possible.
Hypothesis 2: More blood exposures accelerated HIV transmission
In the mid-1980s, Seale proposed that ‘medically promiscuous hypodermics’ were responsible for the emergence of the HIV epidemic.[xxv] In 2000, Chitnis and co-authors identified ‘medical campaigns against smallpox and sleeping sickness’ in French colonies in Central Africa as possible factors in the origin of HIV.[xxvi] Many others have suggested that blood exposures may have played a role.
Injection practices in Europe and the US before 1950
The first syringes with needles small enough to penetrate the skin are credited to Charles Pravaz of France and Alexander Wood of Scotland around 1850. Over time, doctors found more things to inject. Because of insufficient care taken to sterilize instruments between patients, injections exposed people to small amounts of blood – and bloodborne pathogens – from previous patients. This was an unprecedented challenge to human immune systems,[xxvii] which had evolved with reliance on skin to prevent all but rare blood exposures.
In 1883-84, 191 of 1,350 workers at a German shipyard developed jaundice, a symptom of liver disease, after smallpox vaccination.[xxviii] These were the first reported cases of jaundice linked to invasive healthcare. Through the 1940s, doctors in Europe and the US recognized jaundice – likely caused by hepatitis B virus infection – after a variety of invasive procedures including collecting blood and injecting gold salts and other drugs.[xxix]
In the first half of the 20th century – before penicillin – common treatment for syphilis entailed 15 or more weekly intravenous (into a vein) injections of arsenic compounds. In busy clinics, the usual practice was to change needles between patients, but to rinse and reuse syringes. For decades doctors wondered why so many patients treated for syphilis developed jaundice. Many doctors blamed the injected arsenic. Finally, during World War II, doctors working with British and US soldiers were able to show that jaundice was often caused by an (unseen) infectious agent in blood. In the worst single outbreak, 28,000 US soldiers developed jaundice after injections of yellow fever vaccine prepared from infected human sera.[xxx] Around the same time, 68 percent of men treated for syphilis in one clinic in the United Kingdom (UK) developed jaundice.[xxxi]
In a classic 1943-45 study, Laird demonstrated that using a sterile syringe and needle for every injection prevented jaundice after treatment for syphilis. After introducing these procedures in a UK military clinic, he found only one case of post-treatment jaundice in 167 patients.[xxxii] Other research reported in 1943 showed that rinsing syringes did not stop patient-to-patient transmission of bloodborne pathogens. To mimic common procedures during intravenous injections, researchers drew a small amount of blood contaminated with bacteria into a syringe. Next, they rinsed the syringe – drawing in and expelling water – as may as six times. Even after six rinses, they found bacteria in expelled water. The study recommended ‘The use of a freshly boiled syringe for each patient...’[xxxiii]
At the time, a common technique for administering intramuscular (into a muscle) injections, recommended by the UK’s Medical Research Council in 1945, ‘employs a separate sterile needle for each injection, but does not sterilize the syringe, which contains several doses of inoculum, between injections.’[xxxiv] Mid-century research showed this practice also spreads infections. In a study reported in 1950, researchers took syringes containing a sterile solution, injected some of the solution into mice infected with Streptococcus pneumoniae (a bloodborne bacterium), changed the needles, and then injected healthy mice. In two experiments, 35 of 46 previously healthy mice contracted fatal streptococcus infections.[xxxv] Changing needles is not enough to protect subsequent patients, because blood and pathogens reach syringes through suction when needles are removed from the syringes, through back pressure from tissues during injections, and through mixing and movement of fluid through needles.[xxxvi]
In Europe and the US, unsafe practices for injections and other invasive healthcare spread bloodborne pathogens on a prodigious scale during the first half of the 20th century. What happened in Africa?
Injections and other invasive healthcare in Central and West Africa, 1900-60
Due in part to tropical diseases that killed a large percentage of Europeans who ventured inland into Central and West Africa before 1900, Europeans did not claim control over much of the area until the late 19th century. Colonial rule ended in most countries of the region around 1960. Between these dates, colonial governments and missionaries extended European medical systems and techniques into Central and West Africa.
Common invasive procedures included immunizations against smallpox, yellow fever, and tuberculosis (with Bacillus Calmette-Guerin [BCG] vaccine). Common invasive procedures in curative care included injections of quinine to treat malaria, arsenic-based drugs, bismuth salts, and mercury salts to treat syphilis, various medicines to treat leprosy, and after 1950, as many as 60 injections of streptomycin to treat tuberculosis. Long-running programs to control sleeping sickness and yaws may have been the most important public health programs for passage of bloodborne pathogens.
Sleeping sickness control
Europeans recognized sleeping sickness, identified by lethargy and emaciation leading to death, in small numbers of West Africans from the 18th century. In 1901, Forde and Dutton discovered trypanosomes – a one-celled parasite – in the blood of a patient suffering from sleeping sickness in the Gambia.[xxxvii]
Around 1900, the worst recorded epidemic of sleeping sickness erupted east of DRC in Uganda, at that time a British colony. During 1900-20, this epidemic killed an estimated 250,000 people – a third of the population – over 10,000 square kilometers along the north shore of Lake Victoria and extending inland.[xxxviii] At that time and for many decades, most people thought that the trypanosomes that caused the Ugandan epidemic were similar to those that circulate in West and Central Africa. However, this was likely a case of mistaken identity. A recent review of the symptoms of those who died attributes the Ugandan epidemic to Trypansoma brucei, subspecies rhodesiense, which circulates among cattle and wild animals in East and Southern Africa.[xxxix] This sub-species occasionally infects humans, causing local epidemics with rapidly progressing and fatal infections.
The Ugandan epidemic motivated colonial governments to look for trypansomes in their African subjects.[xli] Soon, surveys in the Belgian Congo[xlii] and elsewhere in Central Africa found trypanosomes in small (and rarely large) percentages of Africans in communities exposed to tsetse flies. Colonial doctors feared these infections were the prelude to massive and deadly epidemics. But were they? The trypanosomes that circulate in West and Central Africa, Trypanosoma brucei, subspecies gambiense, infect mostly humans. Many people infected with this sub-species report no or only mild symptoms,[xl] and no studies show how often and how fast untreated infections progress to debilitating illness. For both subspecies, the tsetse fly is an intermediate host, passing trypanosomes among animals and humans.
By 1906, European chemists had developed an injectable arsenic-based drug, atoxyl, that could eliminate trypanosomes from blood. Other drugs followed. With these drugs, doctors in French and Belgian colonies in Central and later West Africa undertook the stupendous task of finding all people with trypanosomes – whether or not they felt sick – and injecting drugs to ‘sterilize’ their blood. The guiding goal was not to cure those who were infected, but to protect others by removing the risk that a biting fly might pick up and pass trypanosomes to them.
In 1917-19, Eugene Jamot, a leader in French efforts to eliminate trypanosomes from African blood, organized a team that toured parts of what is now CAR to find people with trypanosomes, and to sterilize their blood.[xliii] The use of mobile teams for sleeping sickness control spread throughout Central and later West Africa. Visiting teams summoned all adults and children for inspection, routinely relying on militia and on threats to collect people.[xliv] Doctors and assistants examined people for clinical signs of infection, and then looked for trypanosomes in fluid aspirated (taken by syringe and needle) from lymph glands and/or in drops of blood from pricked fingers. Beginning in the 1920s, medics routinely examined the cerebrospinal fluid (extracted by lumbar puncture) of persons with confirmed or suspected infection.
In some communities in southern Cameroon in the late 1920s, mobile teams found trypanosomes in more than 50 percent of the population.[xlv] Through 1932, health officials in Cameroon recorded more than 300,000 people – 12 percent of the population – with trypanosomes. In DRC, mobile teams visited at least a third of the population during most years from 1926 to 1945 and found a cumulative total of 360,000 persons – approximately 5 percent of the population – with trypanosomes.[xlvi] In French West Africa (currently Benin, Burkina Faso, Cote d’Ivoire, Guinea, Mali, Mauritania, Niger, and Senegal), surveys during 1934-42 discovered 20,000-30,000 new cases each year (except 1939, when World War II disrupted sleeping sickness surveys).[xlvii]
Treatment varied from place to place and over time. In 1928, standard treatment in Cameroon was a series of six fortnightly subcutaneous (under the skin) injections of atoxyl or 10 intravenous injections of another arsenic-based drug.[xlviii] Standard treatment in one part of DRC in 1931 was 40 injections administered on consecutive days.[xlix] After a course of injections, doctors repeated tests to assess the success of the treatment. For example, doctors in Kasai Province of DRC in the 1930s followed people for five years after treatment. Not until patients’ cerebrospinal fluid was normal in five consecutive annual lumbar punctures were they considered to be cured.[l] If not, they received another series of injections, and so on. A review of 3,705 persons treated during 1919-1935 in CAR reported a total of 108,003 injections – an average of 29 per patient.[li] One unfortunate patient in DRC in the 1930s received 150 injections in less than five years, along with multiple lumbar punctures.[lii]
New infections and treatment fell off from 1930 in Central Africa, and from the 1940s in West Africa. From the late 1940s, doctors in French and Belgian colonies in Central and West Africa introduced twice-yearly injections of diamidine for prophylaxis against infection by trypanosomes.[liii] These injections concentrated in communities considered to have the greatest risk for sleeping sickness, which generally meant the same communities that had been most intensively treated to date.
Did programs to ‘sterilize’ African blood stop deadly epidemics, or were T.b. gambiense infections in Central and West Africa self-limiting in most individuals and in their communities? In 1938, an official in Nigeria’s Sleeping Sickness Service assessed: ‘Taken as a whole the disease is of a mild type.’ Most people found with trypanosomes had only intermittent symptoms, and had reached a stage where their ‘disease and…resistance to it seem to have obtained a state of equilibrium.’[liv] A 2000-02 study of a suspected sleeping sickness outbreak in a community in DRC supports this less alarmist view. The study found 77 people with trypanosomes, and attributed four deaths to sleeping sickness over three years – showing an average of just over one death per year against 77 infections.[lv] The low ratio of deaths to infections could be explained by a sudden upsurge of infections and effective treatment on the one hand, or by a lot of chronic and non-progressing or defeated infections on the other. The natural course of T.b. gambiense infection in humans is not yet well known.
Yaws commonly presents as persistent sores, and may also damage cartilage and bone. The disease is caused by a one-celled parasite, Treponema pallidum sub-species pertenue, which is closely related to the parasite that causes syphilis. Yaws transmits through non-sexual skin contact, often among children.
In forested regions of southern Cameroon, doctors during the 1940s often found more than 20 percent of the residents with new cases each year, while annual incidence was less than 1 percent in dryer regions further north. A similar distribution of yaws was found in other French colonies in Central Africa – for example, high incidence in Gabon’s forests, but low incidence in Chad’s savannas. During most years from 1936 through the 1950s, doctors annually diagnosed more than 100,000 new cases of yaws in Cameroon, and more than 80,000 in other French colonies in Central Africa. French colonial health services treated yaws with a series of 3-15 intravenous or intramuscular injections of various arsenical or other drugs.
Nosocomial transmission of hepatitis B and C viruses in Africa
In Africa, Jamot recommended boiling syringes and needles for at least 30 minutes, and changing needles for each patient.[lvii] Even if injectors followed these procedures, they would have routinely reused syringes without sterilization. Syringe reuse continued in Africa even after it was recognized to be dangerous in the mid-1940s. For example, a doctor in DRC reported in 1953 that the country[lviii]
is strewn with various medical facilities – maternity clinics, hospitals, dispensaries – where local health care staff daily administer dozens and even hundreds of injections in conditions which make it impossible to sterilize the syringe or needle after each use. At the Red Cross clinic to treat sexually transmitted disease in Leopoldville [Kinshasa] around 300 injections are given each day...The used syringes are simply rinsed, first in water, then in alcohol and ether, and are then ready for reuse…Syringes pass therefore from one patient to another, conserving, sometimes, small quantities of contagious blood…
In Europe and the US, similar practices before 1950 have been retrospectively credited with causing thousands of cases of post-treatment jaundice. What evidence is there that injections before 1960 spread bloodborne pathogens in Africa?
Aside from HIV, hepatitis B and C viruses are the most important bloodborne pathogens recognized to date. Both have circulated among humans in Africa for thousands of years.[lix] However, the percentage of Africans with hepatitis B or C infections appears to have been low until at least 1850. The evidence for this is that there is much more genetic diversity among these viruses in Central and West Africa than in the Americas. This implies that the slave trade during 1500-1850 did not transport many infected Africans to the Americas, which in turn implies that low percentages of Africans were infected before 1850.
Not long after the hepatitis B virus was discovered in the 1960s, surveys found that 8-20 percent of African adults had active (mostly chronic) hepatitis B infections.[lx] In contrast, only 0.2-0.5 percent of people in North America and Western Europe have active hepatitis B infections. The implicit huge increase in hepatitis B infections among Africans after 1850 – if, indeed, that is what happened – could be explained by blood exposures during colonial and post-colonial healthcare. (Chapters 3 and 8 consider the influential view that unsafe healthcare contributes little or nothing to Africa’s hepatitis B problem.)
The hepatitis C virus was discovered in 1989. Just as for hepatitis B, prevalence of hepatitis C infection in Central and West Africa appears to have increased from low levels before 1850 to some of the highest levels in the world. Recent estimates show 13.8 percent of people in Cameroon infected with hepatitis C, 5.5 percent in DRC, and 9.2 percent in Gabon.[lxi] In contrast, in rich countries, the prevalence of hepatitis C infection seldom exceeds 2 percent, and most infections are from IDU or from transfusion of blood or blood products before reliable tests for the virus were introduced in the early 1990s. (Most estimates of the number of people with hepatitis C infections are based on tests for antibodies, not for the virus. Because some people with or without antibodies have defeated the virus, these estimates may overstate current infections, but may also miss many past and defeated infections.)
Infection with hepatitis C points to blood exposures. The virus seldom passes through vaginal or even anal sex, or from mother to child. For example, one study in Italy followed 776 couples in which one partner was infected. After 10 years, with an average of 1.8 coital acts per week, and no condom use – a total of 700,000 unprotected coital acts – not one spouse had acquired hepatitis C from their infected partner.[lxii]
Egypt has the highest estimated prevalence of hepatitis C infection – 18 per cent – in the world.[lxiii] Egypt’s high prevalence has been attributed to public health campaigns in the period 1961-86 to treat schistosomiasis (a parasite infection) with 12-16 intravenous injections.[lxiv] Apparently, many of the injections were delivered with needles and/or syringes reused without sterilization.
In former French and Belgian colonies in Central Africa, several pieces of evidence suggest that high prevalence of hepatitis C infection can be traced to healthcare during colonial rule. Because most people infected with hepatitis C live normal lives – only a minority develops liver disease – one would expect to see a steady increase in infection with age. Older people, just by living longer, have had more blood exposures. However, in Central Africa, and especially in Cameroon, much higher prevalence in persons born before 1960 suggests that rates of transmission of hepatitis C virus were high before 1960, and then fell sharply around 1960 and stayed low for decades.[lxv] For example, a survey in a rural forest area in southern Cameroon in 1990 found that 98 (33 percent) of 298 persons over 40 years old – who were born before 1950 – carried antibodies to hepatitis C, compared to only three of 509 persons aged less than 40 years.[lxvi] An analysis of genetic sequences from several hundred hepatitis C viruses collected in Cameroon concluded that the number of infections began to grow rapidly in 1920, and continued to do so up to 1960.[lxvii]
The colonial response to sleeping sickness – injections to treat and prevent, lumbar punctures and other diagnostic procedures – likely contributed to hepatitis C transmission, especially during the 1920s. At the country level, hepatitis C prevalence is greater in former French colonies (for example, 13.8 percent in Cameroon) than in adjacent former British colonies (for example, 2.1 percent in Nigeria),[lxviii] where authorities paid less attention to sleeping sickness. In Cameroon, injections to treat sleeping sickness peaked over 500,000 per year in the late 1920s before falling dramatically.
On the other hand, injections to treat treponemal infections – primarily yaws, but also syphilis – likely exceeded 1,000,000 per year at times from the mid-1930s to the late 1950s. Similar trends occurred in other French colonies in Central Africa. From this evidence, Pepin and Labbe link injections for yaws and to a lesser extent syphilis to hepatitis C:[lxix]
…in some regions, the whole population developed yaws within a few years… [Y]aws was more common among children who had the opportunity to survive until the mid-1990s… [T]he geographic distribution of HCV [hepatitis C virus] closely corresponds to the historical distribution of yaws whose incidence was much higher in coastal and forested regions…than in northern savannas.’
Tsetse flies, sleeping sickness, yaws, and hepatitis C all concentrated in forested regions. For example, one study in southern Cameroon found that prevalence of hepatitis C antibodies was three times greater in villages surrounded by forests than in villages surrounded by fields.[lxx]
Evidence suggesting that healthcare transmitted HIV
During the period 1900-60, healthcare programs in Central and West Africa administered millions of injections for yaws, syphilis, sleeping sickness, and other diseases. The best estimates for the time and place for the origin of the HIV-1 O and M groups are in Cameroon in the several decades before 1930, when colonial healthcare treated hundreds of thousands of people for sleeping sickness and other diseases. Similarly, SIV apparently passed from sooty mangabeys to begin the HIV-2 A and B groups in or near Cote d’Ivoire sometime before 1940. Through 1945, sleeping sickness surveillance found more than 46,000 people infected with trypanosomes in Cote d’Ivoire.[lxxi]
Africans who traveled, such as civil servants, soldiers, and labor conscripts, received special attention and very likely more injections than others. French and Belgian colonial governments controlled internal travel, establishing filter posts along major transportation arteries to inspect travelers and their health documents. In 1910, for example, the Belgian Congo introduced medical passports.[lxxii] Health posts routinely detained and treated travelers before allowing them to proceed. Coincidences in time and place between colonial health care and the early spread of HIV-1 and HIV-2 do not prove causation, but they allow it.
Because people can live for many years with HIV-2 infections, a study in 2005 among people aged 50 years and older in Guinea-Bissau has been able to identify long-ago blood exposures – some of which occurred before 1960 – as risks for HIV-2 infection. Over half of those tested were aged over 60 years, and therefore born before 1945. People with HIV-2 were more likely than people without HIV-2 to have received injections to treat tuberculosis or sleeping sickness (the peak in sleeping sickness diagnoses in Guinea-Bissau occurred in the early 1950s). Women with HIV-2 infections were more likely to have been circumcised (female circumcision was commonly performed on groups of girls aged 8-12 years). The study team concluded that HIV-2[lxxiii]
spread parenterally [though blood exposures] starting in the late 1940s, through needles, syringes, and instruments used for excision… We documented three routes for parenteral transmission, but it is possible that others existed.
People infected with HIV-1 characteristically have much more virus per cubic centimeter of blood than do people infected with HIV-2. Thus, wherever HIV-1 was available, invasive procedures similar to those that transmitted HIV-2 in Guinea-Bissau likely passed HIV-1 from person-to-person. High prevalence of hepatitis C infection in Central Africa attests to frequent blood exposures during 1920-60. As Pepin and Labbe note, “the same procedures [that spread hepatitis C] could have exponentially amplified HIV-1, from a single hunter/cook occupationally infected…to several thousand patients treated with arsenicals or other drugs…”[lxxiv]
The apparent slowing of hepatitis C transmission in Cameroon and in other former French colonies in Central Africa around 1960 coincides with the end of colonial rule, more effective and shorter treatment for yaws and syphilis (with penicillin replacing arsenicals and bismuth), and a cut-back in sleeping sickness programs. Intriguingly, these changes may have reduced bloodborne transmission of HIV as well. If so, this could help to explain the low prevalence of HIV infection found in Cameroon in the early 1980s (see Chapter 5).
Hypothesis 3: More blood exposures allowed HIV to adapt to humans
One variant of the hypothesis that invasive healthcare in Africa accounts for the emergence of HIV epidemics in the 20th century supposes that it allowed HIV to adapt to humans.[lxxv] According to this hypothesis, the transmission of SIV from a simian to a human leads to an infection that the human can suppress or even eliminate within weeks to months. The hypothesis goes on to propose that blood exposures during healthcare in the 20th century transmitted a rare and fleeting HIV infection through several people, and that this ‘serial passaging’ of HIV though one person after another allowed the virus to live and multiply in humans long enough to adapt. Having adapted, it could maintain sustained infections, and could transmit more easily from one human to another. Serial passaging could have occurred through invasive healthcare in French and Belgian colonies before World War II.[lxxvi]
But was serial passaging necessary? The hypothesis assumes that SIV from chimpanzees would have difficulty living in humans. But many viruses survive and spread after cross-species transmission.
If unsterile invasive procedures were sufficiently common to passage several rare, unadapted HIVs from cut hunters or butchers through 3-4 people in a year, that same frequency of unsterile procedures could spread HIV from one infection to thousands over several decades. To create the HIV epidemic, there would be no need to hypothesize that serial passaging allowed HIV to adapt. HIV has no doubt evolved in humans over time, but the evidence to date does not show that change in HIV (as against change in people’s sexual and/or medical practices) was the key that allowed HIV to spread among humans.
During the colonial era, which ended for most Africans around 1960, a lot of evidence suggests that blood exposures during healthcare spread bloodborne viruses. At least until 1950, most doctors were not aware of how easy it was to transmit infections through instruments reused without sterilization. Hence, much of the damage could be attributed to hubris – doctors thinking they knew more than they did, and thus subjecting patients to unknown risks.
Colonial healthcare programs in Central and West Africa no doubt transmitted HIV along with other bloodborne pathogens. But was this a minor or a major contributor to the emergence of HIV-1 and HIV-2 epidemics? More information from sequencing – especially if old HIV can be sequenced from stored blood – and from other sources may advance our understanding of the early epidemic, and could change current views, as described in this chapter.
[i] Switzer WM, Parekh B, Shanmugam V, et al. ‘The epidemiology of simian immunodeficiency virus infection in a large number of wild- and captive-born chimpanzees: Evidence for a recent introduction following chimpanzee divergence’, AIDS Res Hum Retroviruses, 2005, 21: 335-42.
[ii] Keele BF, Van Heuverswyn F, Li Y, et al. ‘Chimpanzee reservoirs of pandemic and nonpandemic HIV-1’, Science, 2006, 313: 523-6.
[iii] Van Heuverswyn F, Li Y, Neel C, et al. ‘SIV infection in wild gorillas’, Nature, 2006, 444: 164.
[iv] Yusim K, Peeters M, Pybus OG, et al. ‘Using human immunodeficiency virus type 1 sequences to infer historical features of the acquired immune deficiency syndrome epidemic and human immunodeficiency virus evolution’, Phil Trans R Soc Lond B, 2001, 356: 855-66; Korber B, Muldoon M, Theiler J, et al. ‘Timing the ancestor of the HIV-1 pandemic strains’, Science 2000; 288; 1789-96; Salemi M, Strimmer K, Hall WW, et al. ‘Dating the common ancestor of SIVcpz and HIV-1 M group and the origin of the HIV-1 sub-types by using a new method to uncover clock-like molecular evolution’, FASEB J, 2001; 15: 276-8. Worobey M, Gemmel M, Teuwen DE, et al. ‘Direct evidence of extensive diversity of HIV-1 in Kinshasa by 1960’, Nature, 2008, 455: 661-64.
[v] Roques P, Robertson DL, Souquiere S, et al. ‘Phylogenetic analysis of 49 newly derived HIV-1 group O strains: high viral diversity but no group M-like subtype structure’, Virology, 2002, 302: 259-73.
[vi] Lemey P, Pybus OG, Rambaut A, et al. ‘The molecular population genetics of HIV-1 group O’, Genetics, 2004, 167: 1059-68.
[vii] Roques P, Robertson DL, Souquiere S, et al. ‘Phylogenetic characteristics of three new HIV-1 N strains and implications for the origin of group N’, AIDS, 2004, 18: 1371-81.
[viii] Zhu T, Korber BT, Nahmias AJ, et al. ‘An African HIV-1 sequence from 1959 and implications for the origin of the epidemic’, Nature, 1998, 391: 594-7; Motulsky AG, Vandepitte J, Fraser GR. ‘Population genetic studies in the Congo: I. Glucose-6-phosphate dehydrogenase deficiency, hemoglobin S and malaria’, Am J Hum Genet, 1966, 18: 514-37.
[ix] Worobey M, Gemmel M, Teuwen DE, et al. ‘Direct evidence of extensive diversity of HIV-1 in Kinshasa by 1960’.
[x] Hooper E. The River. London: Penguin, 2000. p. 260.
[xi] Sonnet J, Michaux J-L, Zech F, et al. ‘Early AIDS cases originating from Zaire and Burundi (1962-1976)’, Scand J Infect Dis, 1987, 19: 511-17.
[xii] Vangroenweghe D. ‘The earliest cases of human immunodeficiency virus type 1 group M in Congo-Kinshasa, Rwanda and Burundi and the origin of acquired immune deficiency syndrome’, Phil Trans R Soc Lond B, 2001, 356: 923-5; Jonassen TO, Stene-Johansen K, Berg ES, et al. ‘Sequence analysis of HIV-1 group O from Norwegian patients infected in the 1960s’, Virology, 1997, 231: 43-7.
[xiii] Thijs A. ‘L’angiosarcomatose de Kaposi au Congo belge et au Ruanda-Urundi’, Ann Soc Belge Med Trop, 1957, 37: 295-305.
[xiv] Molez J-F. ‘The historical question of acquired immunodeficiency syndrome in the 1960s in the Congo River basin area in relation to cryptococcal meningitis’, Am J Trop Med Hyg, 1998, 58: 273-6; Devreese A, Donkers J, Ninane G, et al. ‘Histoplasmose africaine a formes capsulatum causee par Histoplasma dubosii Vanbreuseghem 1952’, Ann Soc Belge Med Trop, 1961, 5: 403-14.
[xv] Los Alamos National Laboratory. HIV Sequence Database. Available at: http://www.hiv.lanl.gov/content/hiv-db (accessed 28 September 2006).
[xvi] Lemey P et al. ‘The molecular population genetics of HIV-1 group O’.
[xvii] Smith SM, Christian D, de Lame V, et al. ‘Isolation of a new HIV-2 group in the US’, Retrovirology, 2008, 5: 103.
[xviii] Santiago ML, Range F, Keele BF, et al. ‘Simian immunodeficiency virus infection in free-ranging sooty mangebeys (Cercocebus atys atys) from the Tai Forest, Cote d’Ivoire: implications for the origin of endemic human immunodeficiency virus type 2’, J Virol, 2005, 79: 12515-27.
[xix] Lemey P, Pybus OG, Wang B, et al. ‘Tracing the origin and history of the HIV-2 epidemic’, Proc Nat Acad Sci USA, 2003, 100: 6588-92; Personal communication from Philippe Lemey, 28 September 2006.
[xx] Kawamura M, Yamazaki S, Ishikawa K, et al. ‘HIV-2 in West Africa in 1966 [letter]’, Lancet, 1989, i: 385; Le Guenno B. ‘HIV1 and HIV2: Two ancient viruses for a new disease’, Trans Roy Soc Trop Med Hygiene, 1989, 83: 847.
[xxi] Wilkins A, Ricard D, Todd J, et al. ‘The epidemiology of HIV infection in a rural area of Guinea-Bissau’, AIDS, 1993, 7: 1119-22; Piedade J, Venenno T, Prieto E, et al. ‘Longstanding presence of HIV-2 infection in Guinea-Bissau (West Africa)’, Acta Trop, 2000, 76: 119-24; Pepin J, Plamondon M, Alves AC, et al. ‘Parenteral transmission during excision and treatment of tuberculosis and trypanosomiasis may be responsible for the HIV-2 epidemic in Guinea-Bissau’, AIDS, 2006, 20: 1303-11.
[xxii] Wolfe ND, Switzer WM, Carr JK, et al. ‘Naturally acquired simian retrovirus infections in central African hunters’, Lancet, 2004, 363: 932-7.
[xxiii] Hooper E. The River.
[xxiv] Seale JR, Medvedev ZA. ‘Origin and transmission of AIDS. Multi-use hypodermics and the threat to the Soviet Union: Discussion paper’, J Roy Soc Med, 1987, 80: 301-4. p. 302.
[xxv] Ibid., p. 302.
[xxvi] Chitnis A, Rawls D, Moore J. ‘Origin of HIV type 1 in colonial French Equatorial Africa?’ AIDS Res Hum Retroviruses, 2000, 16: 5-8. p. 7.
[xxvii] Drucker E, Alcabes PG, Marx PA. ‘The injection century: Massive unsterile injections and the emergence of human pathogens’, Lancet, 2001, 358: 1989-92.
[xxviii] Dull HB, ‘Syringe-transmitted hepatitis: A recent epidemic in historical perspective’, JAMA, 1961, 176: 413-8.
[xxix] Laird SM. ‘Syringe-transmitted hepatitis’, Glasgow Med J, 1947, 28: 199-219.
[xxx] ‘Jaundice following yellow fever vaccination’, JAMA, 1942, 119: 1110.
[xxxi] Salaman MH, King AJ, Williams DI, et al. ‘Prevention of jaundice resulting from antisyphilitic treatment’, Lancet, 1944, ii: 7-8.
[xxxii] Laird SM. ‘Syringe-transmitted hepatitis’.
[xxxiii] Bigger JW. ‘Jaundice in syphilitics under treatment’, Lancet, 1943: i: 457-8. p. 458.
[xxxiv] Evans RJ, Spooner ETC. A possible mode of transfer of infection by syringes used for mass inoculation. Br Med J, 1950, ii: 185-8. p. 185.
[xxxvi] Hughes RR, Post-penicillin jaundice, Br Med J, 1946, 2: 685-8.
[xxxvii] Boyce R, Ross R, Sherrington CS. ‘Note on the discovery of the human trypanosome [letter]’, Brit Med J, 1902, 2: 1680.
[xxxviii] Christy C. ‘Sleeping sickness’, Journal of the Royal African Society, 1903, 9(3): 1-11; Low GL. ‘A retrospect of tropical medicine from 1894 to 1914’, Trans R Soc Trop Med Hyg, 1929, 23: 213-34; Maudlin I. ‘African trypanosomiasis’, Ann Trop Med Parasitology, 2006, 100: 679-701.
[xxxix] Fevre EM, Coleman PG, Welburn SC, et al. ‘Reanalyzing the 1900-1920 sleeping sickness epidemic in Uganda’, Emerg Infect Dis, 2004, 10: 567-73.
[xl] Lester HMO. ‘Further progress in the control of sleeping sickness in Nigeria’, Trans Roy Soc Trop Med Hygiene, 1945, 38: 425-44.
[xli] Lyons M. ‘From ‘death camps’ to condon sanitaire: the development of sleeping sickness policy in the Uele district of the Belgian Congo, 1903-1914’, Journal of African History, 1985, 26: 69-91.
[xlii] Dutton JE, Todd JL. ‘Distribution and spread of sleeping sickness in the Congo Free State’, in: Dutton JE, Todd JL. Reports of the Expedition to the Congo, 1903-5, Liverpool School of Tropical Medicine – Memoir XVIII. London: Williams & Norgate, 1906. pp 23-38.
[xliii] Richet P. ‘Eugene Jamot: Son oeuvre’, Med Trop 1979; 39: 487-93.
[xliv] Headrick R. Colonialism, Health and Illness in French Equatorial Africa, 1885-1935 (Headrick DR, ed). Atlanta: Africa Studies Association, 1994.
[xlv] Jamot E. ‘La maladie du sommeil au Cameroun en janvier 1929’, Bull Soc Pathol Exot,1929; 22: 473-96.
[xlvi] Van Hoof LMJJ. ‘Observations on trypanosomiasis in the Belgian Congo’, Trans R Soc Trop Med Hyg, 1947, 40: 728-61.
[xlvii] Masseguin A, Taillefer-Grimaldi J. ‘Declin et danger residuel de la trypanosomiase en Africque Occidentale Francaise’, Ann Soc Belg Med Trop, 1954, 34: 671-96.
[xlviii] Hermant I. ‘Les maladies transmissibles observees dan les colonies francaises et territories sous mandate pendant l’annee 1928’, Ann Med Pharmacie Coloniales, 1931, 29: 5-138.
[xlix] Lyons M. The Colonial Disease: A social history of sleeping sickness in northern Zaire, 1900-1940. Cambridge: Cambridge University Press, 1992. p. 152.
[l] Haveaux G. ‘Vingt ans d’action medicale contre la maladie du someil dans le Kasai’, Ann Soc Belge Med Trop, 1945, 25: 155-203.
[li] Vamos S. ‘Traitment de trypansomes dans un secteur du Moyen-Chari (A.E.F.) etude de 3,705 observations’, Bull Soc Pathol Exot, 1936, 29: 1015-22.
[lii] Van Hoof L, Henrard C, Peel E. ‘Contribution a l’epidemiologie de la maladie du sommeil au Congo Belge’, Ann Soc Belge Med Trop, 1938, 18: 143-201.
[liii] Lotte A. ‘Enseigment de quatre annees de chimio-prophylaxie en A.E.F.’, Med Trop 1951, 11: 737-66; Deroover J. ‘Modifications de l’aspect de la trypanosomiase humaine a Tr. gambianse dans un vieux foyer, sous l’influence des methods modernes de prophylaxis et de therapeutique’, Ann Soc Belge Med Trop Trop, 1958, 38: 149-78.
[liv] Lester HMO. ‘The progress of sleeping sickness work in northern Nigeria’, West Afr Med J, 1938, 10: 2-10. p. 2.
[lv] Lutumba P, Makieya E, Shaw A, et al. ‘Human African trypanosomiasis in a rural community, Democratic Republic of Congo’, Emerg Infect Dis, 2007, 13: 248-54.
[lvi] Pepin J, Labbe A-C. ‘Noble goals, unforeseen consequences: control of tropical diseases in colonial Central Africa and the iatrogenic transmission of blood-borne viruses’, Trop Med Int Health, 2008, 13: 744-753.
[lvii] Jamot E. ‘La maladie du sommeil’.
[lviii] Beheyt P. ‘Contribution a l’etude des hepatites en Afrique: L’hepatite epidemique et l’hepatite par inoculation’, Ann Soc Belge Med Trop, 1953, 33: 297-338. p. 335. Gisselquist translated the quote.
[lix] Simmonds P. ‘Reconstructing the origins of human hepatitis viruses’, Phil Trans R Soc Lond B, 2001, 356: 1013-26.
[lx] Kiire CF. ‘The epidemiology and control of hepatitis B in sub-Saharan Africa’, Prog Med Virol, 1993, 40: 141-56; WHO. Hepatitis B. Geneva: WHO, 2002. Doc. no: WHO/CDS/CSR/LYO/2002.2:Hepatitis B.
[lxi] Madhava V, Burgess C, Drucker E. ‘Epidemiology of chronic hepatitis C virus infection in sub-Saharan Africa’, Lancet Infect Dis, 2002, 2: 293-302.
[lxii] Vandelli C, Renzo F, Romano L, et al. ‘Lack of evidence of sexual transmission of hepatitis C among monogamous couples: Results of a 10-year prospective follow-up study’, Am J Gastroenterol, 2004, 99: 855-9.
[lxiii] ‘Hepatitis C – Global prevalence (update)’, Wkly Epidemiol Rec, 1999; 74: 425-7.
[lxiv] Frank C, Mohamed MK, Strickland GT, et al. ‘The role of parenteral antischistomal therapy in the spread of hepatitis C virus in Egypt’, Lancet, 2000, 355: 887-91.
[lxv] Nerrienet E, Pouillot R, Lachenal G, et al. ‘Hepatitis C virus infection in Cameroon: A cohort effect’, J Med Virol, 2005, 76: 208-14.
[lxvi] Louis FJ, Maubert B, Le Hesran JY, et al. ‘High prevalence of anti-hepatitis C virus antibodies in a Cameroon rural forest area’, Trans R Soc Trop Med Hyg, 1994, 88: 53-4.
[lxvii] Njouom R, Nerrienet E, Dubois M, et al. ‘The hepatitis C virus epidemic in Cameroon: genetic evidence for rapid transmission between 1920 and 1960’, Infect Genet Evol, 2007, 7: 361-7. p. 361.
[lxviii] Madhava V, Burgess C, Drucker E. ‘Epidemiology of chronic hepatitis C virus infection’.
[lxix] Pepin J, Labbe A-C. ‘Noble goals, unforeseen consequences: control of tropical diseases in colonial Central Africa and the iatrogenic transmission of blood-borne viruses’. p. 750.
[lxx] Louis FJ, Kemmenge J. ‘Grande variations de la prevalence de l’infection par le virus C des hepatites en Afrique centrale [letter]’, Med Trop, 1994; 54: 277-8.
[lxxi] Masseguin A, Taillefer-Grimaldi J. ‘Declin et danger residuel de la trypanosomiase en Africque Occidentale Francaise’, Ann Soc Belg Med Trop, 1954, 34: 671-96.
[lxxii] Lyons M. The Colonial Disease: A social history of sleeping sickness in northern Zaire, 1900-1940. Cambridge: Cambridge University Press, 1992.
[lxxiii] Pepin J et al. ‘Parenteral transmission’, p. 1309.
[lxxiv] Pepin J, Labbe A-C. ‘Noble goals, unforeseen consequences: control of tropical diseases in colonial Central Africa and the iatrogenic transmission of blood-borne viruses’. p. 751.
[lxxv] Marx P, Alcabes PG, Drucker E. ‘Serial human passage of simian immunodeficiency virus by unsterile injections and the emergence of epidemic human immunodeficiency virus in Africa’, Phil Trans R Soc Lond B. 2001, 356: 911-20.
[lxxvi] Moore J. ‘The puzzling origins of AIDS’, Am Sci, 2004, 92: 540-7.