Epidemic History and Iatrogenic Transmission of Blood-borne Viruses in Mid-20th Century Kinshasa

1. Background

   1.1 The emergence of HIV

   The HIV/AIDS pandemic has so far caused about 29 million deaths, while 33 million individuals currently live with HIV. Even if this will have no direct impact on the future course of the epidemic, it is important to try to understand the factors that allowed the successful emergence of HIV-1, first as a moral obligation towards the victims, and then to draw lessons that could ultimately help mankind avoid facing similar threats in the future.

   There is now no doubt that the source of HIV-1 group M (the strain causing the pandemic) is the Pan troglodytes troglodytes chimpanzee of central Africa. This primate inhabits south Cameroon, Gabon, Equatorial Guinea, the Congo Republic, the south-west of the Central African Republic (CAR), the Cabinda enclave and a small part of the Democratic Republic of Congo (DRC) (the Mayombe area north of the river). It is thought that the original cross-species transmission, from chimpanzee to man, occurred around the beginning of the 20th century, probably through the manipulation of chimpanzee meat by a hunter or a cook, so that the common ancestor of group M existed in humans around 1910-1920. It is extremely unlikely that this very first infected human lived in the DRC, where populations of Pan troglodytes troglodytes were very small. The virus then slowly disseminated along trading routes, eventually reaching Léopoldville and Brazzaville no later than 1959.

   It is in this large bi-national conurbation that the virus managed to flourish and diversify, as supported by several findings:

   1. The broadest genetic diversity of HIV-1 is found in Kinshasa and Brazzaville. All HIV-1 subtypes and many recombinants have been found there. Indeed, the genetic diversity of HIV-1 in Kinshasa in the mid-1980s, twenty-five years ago, was higher than that currently found in any other part of the world.
   2. The two most ancient isolates of HIV-1 have been located in Kinshasa, one from a blood sample obtained in 1959 and the second from a lymph node biopsy obtained in 1960.
   3. Testing of serum collections stored for years showed that among mothers bringing their children to an under-five clinic in the Lemba district of Kinshasa, HIV prevalence was 0.25% in 1970 and 3.0% in 1980.
   4. HIV prevalence among the general adult population of Kinshasa and Brazzaville in the mid-1980s was already between 5 and 8%, while it was far lower in Yaoundé, Douala and Libreville, the other large cities of central Africa.
   5. Several serologically proven cases of AIDS were diagnosed retrospectively among Belgian nationals who were probably infected in the Congo in the 1960s.

   The question that remains is to understand the factors that facilitated the emergence of HIV-1 within this population in the middle of the 20th century (and not elsewhere). For a long time, this has been attributed essentially to the urbanisation of central Africa. The colonialists created cities in which adult men by far outnumbered women, leading to the development of urban prostitution, so that eventually HIV-1 encountered conditions that were propitious for its sexual spread, initially among prostitutes and their clients, along the lines of what was later to be documented in Nairobi in the 1980s.

   It is indeed very likely that prostitution ultimately played a substantial role in the emergence of HIV-1 in Kinshasa/Brazzaville. However, for a long period of time, prostitution in these two cities was rather of a "soft", low-risk, type: free women (ndumbas) had 3-4 regular clients, to whom they provided not just sex but diversified services (cooking, laundry, hair dressing, having a conversation) in exchange for regular financial support rather than a fixed fee per intercourse. High-risk prostitution, in which women could have sex with 3-4 different clients every day, up to a thousand per year, appeared only circa 1960, around new venues called flamingos, bars with rudimentary rooms nearby, in the context of the profound social and economic changes prompted by the country's accession to independence.

   For several years we have been investigating the possibility that some of the early expansion of HIV occurred through well-intentioned medical interventions. For a long time, because the early pharmaceutical products were poorly effective, drugs for the treatment of tropical and other infectious diseases had to be administered intravenously, through syringes and needles that were continuously re-utilised and poorly sterilized, potentially allowing the transmission of blood-borne viruses. This risk could not be appreciated at the time, initially because the existence of such viruses was unknown, but also because acute infection with the Hepatitis C virus (HCV) generally caused only a mild non-specific disease, while most adults had already been infected with the Hepatitis B virus (HBV) during childhood, so that they were immune by the time they received treatments against tropical diseases.

   In Egypt, millions were infected with HCV during interventions for the control of schistosomiasis, demonstrating that iatrogenic epidemics can indeed reach a massive scale. In a community-based study of elderly individuals in Guinea-Bissau, we showed that the HIV-2 epidemic might also have been largely iatrogenic, through the treatment of African trypanosomiasis and the parenteral treatment of tuberculosis with intramuscular streptomycin (augmented by transmission during ritual excision of the clitoris in collective ceremonies), which contributed to the emergence of this other simian-turned-human retrovirus 40 to 50 years ago. This latter study was possible because HIV-2 increases mortality only 2 to 3-fold, in contrast with HIV-1 which enhances mortality 10-fold.

   We hypothesized that the same medical interventions, or others based on the mass administration of parenteral drugs, may also have facilitated, at roughly the same time, the emergence of HIV-1 in central Africa. Because of the high mortality associated with HIV-1 infection, it is not possible to examine this hypothesis directly. However, HCV and HTLV-1 (viruses compatible with a prolonged survival) infections can be used as markers for the parenteral transmission of viruses, to determine indirectly whether those interventions may have contributed to the emergence of HIV-1.

   We then conducted two epidemiological studies of elderly individuals (aged 55-60 years or more) in central Africa. In Ebolowa, south Cameroon, where more than half of the elderly are HCV-seropositive, HCV seropositivity was associated mainly with the intravenous treatment of malaria with quinine, and also with traditional circumcision of boys in group ceremonies. In the Nola region of southwest CAR, HCV seropositivity was associated with the treatment of African trypanosomiasis before 1950, while HTLV-1 infection was associated with prophylactic injections of pentamidine (for the prevention of trypanosomiasis, between 1947 and 1953) and with transfusions.

   In this context, it would be relevant to try to assess whether parenteral transmission might have played a role in the emergence of HIV-1 in Kinshasa half a century ago, using the same two proxies, HCV and HTLV-1 infections.

   1.2 Hepatitis C virus infection in central Africa

   Africa is the continent with the highest HCV prevalence among its adult population. In central Africa, 6% of adults are HCV-seropositive, compared to 2.4% in West Africa and 1.6% in eastern and southern Africa3. In many areas of central Africa, HCV prevalence increases steeply with age, far beyond what could be expected from cumulative exposure over time. Since there is only a modest heterosexual and vertical transmission of HCV, it has generally been thought that most of this transmission occurred parenterally, through injections, transfusions, immunizations and scarifications. It is now known that the parenteral transmission of HCV is rather ineffective unless the needle had first been inserted in the vein of a viremic individual. This was the mechanism that drove the massive epidemic in Egypt, where a large number of individuals were infected in the early 1960s during campaigns for the control of schistosomiasis through the intravenous injections of antimonial drugs.

   In sub-Saharan Africa, Cameroon is the country where the epidemiology of HCV has been most thoroughly investigated. Several studies documented a high HCV prevalence, in Yaoundé but also in many rural southern communities. A synthesis of these studies demonstrated a cohort effect: in many areas, HCV prevalence was 40-50% amongst individuals born before 1945, progressively decreasing to 15% for those born around 1960 and only 3-4% for those born after 1970. In several Cameroonian studies, HCV prevalence peaked in cohorts corresponding to dates of birth around 1935. Furthermore, molecular clock analyses revealed that the number of Cameroonians infected with HCV started increasing exponentially around 1920 for genotype 4, and around 1940 for genotypes 1 and 2. Since other modes of transmission are rather ineffective, this implies a massive iatrogenic transmission of the virus in the last few decades of the colonial era. As mentioned, in one of these high-prevalence communities, the administration of intravenous antimalarials was the main risk factor for HCV infection. It follows that if the HCV blood-borne virus was massively transmitted during medical interventions in several areas inhabited by the Pan troglodytes troglodytes chimpanzee, the same amplification might have occurred with SIVcpz which eventually became HIV-1.

   In the DRC, HCV prevalence has been measured in only one study. Among Kinshasa sex workers, HCV prevalence increased with age and reached 21% in those aged 41-55 years. As the sexual transmission of HCV is rather ineffective, this prevalence must have reflected to some extent the prevalence in the general adult population. Among pregnant women aged 31-55 years, HCV prevalence was only 5.6% but one can presume that few of them were aged more than 40 years.

2. Objectives

Our hypothesis is that the intravenous treatments of various infectious diseases with poorly sterilized syringes and needles contributed to the transmission of HCV and HTLV-1 within the city of Kinshasa in the middle of the 20th century, and that this could have as well facilitated the early emergence of HIV-1.

The objectives of this study will be:

1. to determine whether HCV seropositivity (primary outcome) is associated with past intravenous treatments against tropical and other diseases, especially during the pre-1980 period.
2. to reconstruct the past dynamics of HCV in Kinshasa through molecular clock methods based on the sequencing of local HCV isolates obtained from elderly individuals and determine the period during which maximal transmission of this blood-borne virus occurred.
3. to determine whether the same medical interventions were associated with the transmission of HTLV-1 (secondary outcome).
4. to contribute to a better understanding of the epidemiology of HCV and HTLV-1 infections in the DRC.