VIRAL HAEMORRHAGIC FEVER IN THE SUDAN, 1976 : HUMAN VIROLOGICAL AND SEROLOGICAL STUDIES
E.T.W. BOWEN (1), G.S. PLATT (1), G. LLOYD (1), L.B. McARDELL (1), D.I.H. SIMPSON (2,3), D.H. SMITH (3,4), D.P. FRANCIS (3,5), R.B. HIGHTON (3,4), M. CORNET (3,6), C.C. DRAPER (2,3), BABIKER EL TAHIR (7), ISIAIH MAYOM DENG (7), PACIFICO LOLIK (7), OLIVER DUKU (7)
1. Microbiological Research Establishment, Porton Down, Salisbury SP4 OJG, England.
2. London School of Hygiene and Tropical Medicine.
3. WHO Investigation Team, Sudan.
4. Division of Communicable Diseases, Ministry of Health, Nairobi.
5. Department of Microbiology, Harvard School of Public Health.
6. Institut Pasteur, Dakar.
7. Ministry of Health, Khartoum and Juba, Sudan.
Two further strains of Ebola virus were isolated from acute phase sera collected from acutely ill patients in Maridi hospital during the investigation in November 1976. Antibodies to Ebola virus were detected by immunofluorescence in 42 and 48 patients in Maridi who had been diagnosed clinically but only in six of 31 patients in Nzara. The possibility of the indirect immunofluorescent test not being sufficiently sensitive is discussed.
19% of Maridi case contacts in hospital and in the local community had antibodies. Very few of them gave any history of illness indicating that Ebola virus can cause mild or even subclinical infections. 37% of the cloth room workers in the Nzara cotton factory appeared to have been infected suggesting that the factory may have been the prime source of infection.
An outbreak of viral haemorrhagic fever in Nzara and Maridi townships of southern Sudan in the second half of 1976 affected nearly 300 people, 150 of whom died from the infection. The virus responsible, since named Ebola, was first isolated from patients in October 1976 and was shown to be identical to virus strains isolated during a simultaneous epidemic occurring in northern Zaire (1,2,3).
A brief description of the Sudanese outbreak was outlined by Simpson et al. (4); Smith et al. (5) gave details of the principal clinical features; Francis et al. (6) gave a detailed account of the epidemiological aspects; while Ridley et al. (7) described the histopathological changes found in two fatal cases.
This paper describes additional virus isolation attempts on acute human material and post mortem specimens collected by the joint WHO/Sudanese investigation team and serological studies on human sera collected from convalescent patients, case contacts and members of the local populations of Nzara and Maridi who were not believed to have been exposed to infection with Ebola virus during the outbreak.
MATERIALS AND METHODS
Acute samples: Blood, throat swabs, and urine samples were collected within five days of the onset of illness from eight acutely ill patients in Maridi Hospital isolation wards. All eight patients were febrile, grossly dehydrated and displayed many of the clinical features described by Smith et al. (5) .
In the isolation wards and when handling samples, team members wore protective clothing and biological respirators as suggested by Simpson . Blood samples were collected using disposable syringes and needles and immediately injected into vacutainer tubes. Throat swabs were placed in sterile saline in screw-capped polypropylene serum tubes (Nunc); urine samples were collected from bed-pan samples in similar polypropylene tubes. Specimen containers were washed in 2% formol saline solution, and placed in plastic bags for removal from the isolation area. Swabs, needles and syringes were immersed in 2, formol saline, placed in plastic bags inside a bucket and taken for immediate incineration. Protective clothing was removed on leaving the isolation area; disposable items were incinerated, reusable gowns were decontaminated by boiling.
After clotting, blood was centrifuged in sealed vacutainer tubes in the field laboratory and serum aspirated off into polypropylene tubes. All acute samples were stored in liquid nitrogen prior to dispatch to the Microbiological Research Establishment, Porton.
Post-mortem samples. Small pieces of liver, spleen, heart, lung, kidney and bone marrow were collected in polypropylene tubes and stored in liquid nitrogen.
Convalescent plasma. Aliquots of immune plasma were collected using a modified plasmapheresis technique. As no centrifuge was available, plasma and cells were separated by gravity only over a 2-3 hour period.
Sera from contacts . Blood samples were collected using disposable syringes and needles before centrifugation in sealed vacutainer tubes. In Nzara samples were collected in the cotton factory and in homesteads around the township; in Maridi samples were collected in the hospital grounds, schools and surrounding homesteads. After centrifugation sera were stored in polypropylene tubes and stored at -5ºC. For dispatch they were placed in liquid nitrogen and stored at the Microbiological Research Establishment at -20ºC until tested.
Laboratory studies. All manipulations involving potentially infectious material were carried out in the Maximum Security Laboratory at the Microbiological Research Establishment.
Virus isolation attempts. Ten per cent suspensions of post mortem tissues were prepared in a pH 7.2 phosphate buffered saline containing 0.75% bovine plasma albumin (Armour Fraction V), penicillin and streptomycin. Tissue suspensions, acute serum samples, urine and throat swab eluates were inoculated into cultures of an African green monkey cell line (Vero) and Dunkin-Hartley strain guinea pigs following the procedures described by Bowen et al.
Immunofluorescent Technique. The indirect immunofluorescent test was similar to that described by Wulff and Lange (9) . Vero cells were prepared in 25 cm2 plastic flasks (Nunc) and maintained in Leibovitz medium (LI5) containing 2% feotal calf serum at pH 7.2-7.4. Each flask was inoculated with 1 ml of Ebola virus as a 10% guinea pig liver and spleen suspension. The medium was changed on the second day to reduce toxicity caused by using guinea pig tissue. As indicated by daily immunofluorescence antibody (IFA) testing of replicate preparations, flasks were incubated at 37ºC until 30-40% of the cells had become infected, usually at 6-7 days.
For slide preparation media was first decanted off the infected cells and the cell sheet washed three times in a pH 7.27.4 phosphate buffered saline (PBS). The cell sheet was detached with 2 ml of 0.2% trypsin in versene; the cell suspension was removed and diluted in an equal volume of PBS containing 3% calf serum to inactivate the trypsin/versene activity. The suspension was centrifuged at 1000 rpm for five minutes, the supernatant discarded, and the cells resuspended in PBS. This procedure was repeated twice and the cells were finally suspended in PBS containing 0.2% of bovine serum albumin to a concentration of approximately 5 x 10^5 cells/ml. The suspension was seeded into the wells of polytetrafluoroethylene coated glass slides each having 12 x 6 mm wells. Each well received one drop of cell suspension. The slides were allowed to dry under an ultra-violet lamp for 20 minutes followed by ten minutes fixation in chilled acetone.
Sera were examined for the presence of Ebola virus antibodies by an indirect immunofluorescent method (9). Duplicate dilutions of all test sera were made in PBS and sera were screened at dilutions of 1:4 and 1:8. Known human positive and negative sera were included as controls and PBS controls similarly used. All test and control sera were screened on both infected and uninfected slides. Diluted sera were placed in slide wells and held in a moist chamber at 37ºC for 30 minutes and then washed in PBS for 5-10 minutes. Slides were then dried in air and a drop of Fluorocein labelled rabbit anti-human IgG conjugate (Wellcome) at dilutions of 1:20 to 1:25 applied to all wells. The slides were again held at 37ºC in a moist chamber for 30 minutes and then washed three times in PBS for 2-4 minutes on each occasion and finally rinsed in distilled water for 30 seconds. Slides were then dried in air and examined under a Reichert Fluorvar microscope using an HBO 500w mercury vapour burner. Sera were only accepted as positive if clear fluorescence was observed at dilutions gte 1:8. Sera which were positive only at a dilution of 1:4 were regarded as equivocal and graded as negative.
Virus isolation. Two strains of Ebola virus were isolated from acute human sera from two patients in the Maridi isolation wards - one from a 12-year-old schoolgirl (collected on 5 November 1976) who died three days later; the other, collected on 8 November 1976, was from an 18-year-old male student who later recovered. The isolations were made in guinea pigs inoculated intraperitoneally. The guinea pigs developed a febrile illness five days after inoculation; passage of guinea pig blood to further guinea pigs produced a similar febrile illness but a fatal infection was only produced in guinea pigs after five passages. The isolates were later successfully cultured in Vero cells and identified as strains of Ebola by immunofluorescence.
No isolates were obtained from other patients' sera or from any of the throat swabs, urines or post mortem tissues. Virus particles were, however, visualised by electronmicroscopy in liver sections from both fatal cases.
Serology. One hundred and four surviving Maridi patients were identified. All of them had been diagnosed as suffering from Ebola fever on clinical grounds alone. 35 of these patients were members of Maridi hospital staff and the remaining 69 patients were Maridi residents, some of whom were relatives of hospital staff. 30 of the 35 surviving hospital staff were bled for serological studies. 25 of them (83%) had detectable IFA antibodies. Only 18 of the 69 patients who were not hospital staff were willing to be bled. 17 of these patients had detectable antibodies. Altogether 42 of the 48 patients diagnosed clinically as having been infected and who were bled had antibodies against Ebola virus (Table 1).
Among the probable and possible case contacts hospital staff were considered to be among those most at risk. 159 members of the staff had given no history of severe febrile illness during the epidemic. Sera from 64 of these staff members Were tested for Ebola virus antibodies. The results are shown in Table 2. Seven staff members, including four nurses, a cleaner, a toilet cleaner and a water carrier had evidence of infection. None of the three doctors attending sick patients had antibodies.
One hundred and two members of the Maridi population who had been in contact with known cases of infection were tested. The majority of them were close family contacts and several had helped to nurse sick relatives during their illnesses. 22 case contacts had demonstrable antibodies (Table 3). Close questioning revealed that nine of these close contacts had some evidence of febrile illness without manifesting severe disease. 29 members of a Maridi school were tested as a control group. None of them was thought to have had any contact with a known or suspected case of Ebola virus disease. However, three schoolboys had antibodies. None of them had any history of recent illness.
In Nzara 37 surviving patients diagnosed clinically as having had Ebola fever were identified and 31 of them were bled for antibody studies. Only six (19%) had demonstrable antibodies and none had antibody levels greater than 1:32 (Table 4). Among close family contacts only one person out of 78 who were tested had detectable antibodies although five of these contacts had an antibody titre of 1:4 which not accepted as positive.
RESULTS OF IFA TESTS ON CONVALESCENT PATIENTS' SERA
Number of survivors
Nurses and midwives
RESULTS OF IFA TESTS ON SERA FROM MARIDI HOSPITAL STAFF CONTACTS
Nursing staff (including tutors, midwives, theatre staff)
RESULTS OF IFA TESTS ON SERA FROM MEMBERS OF THE MARIDI POPULATION
No known contact
RESULTS OF IFA TESTS ON SERA FROM NZARA CONVALESCENT PATIENTS AND FAMILY CONTACTS
Number of survivors
Number of contacts bled
Sera from 109 cotton factory workers were tested (Table 5). Seven members of the staff had antibodies with the largest number, three, in the cloth room and adjacent store. Three workers in the weaving section had antibodies while only one member of the 28 spinning section staff who were bled was positive. Antibody levels ranged from 1:16 to 1:64 and none of the seven workers gave a history of any recent illness.
RESULTS OF IFA TESTS ON SERA FROM NZARA COTTON FACTORY STAFF
Cloth room and store
No really satisfactory serological test has yet been devised to detect and quantify antibodies against either Marburg or Ebola viruses. Early workers (11,12,13) used complement fixation tests to detect antibodies to Marburg virus following the outbreak in 1967. They used crude antigens derived from guinea pig and monkey tissues but found them to be unreliable because they lacked specificity. Malherbe and Strickland-Cholmley (14) compared antigens derived from animal tissues with those prepared in monkey kidney cell cultures and found both types to be unsatisfactory. Slenczka and Wolff (15) and Slenczka et al. (16) found that antigens derived from chronically infected Vero cells were much more satisfactory but still had doubts about their activity.
Neutralization tests have been carried out in guinea pigs and tissue culture systems but have proved costly in animals, cumbersome, time consuming and generally very unsatisfactory. Siegert and Slenczka (13) and ourselves have had some success with neutralization tests in Vero and BHK21 cells based on counts of intracytoplasmic inclusion bodies but the results were difficult to reproduce. The problem is that no really obvious cytopathic changes can be readily visualized in any of the cell culture systems used so far although Johnson et al. (2) have claimed that some cytopathic changes are discernible in Vero cells. However these workers, like ourselves, have based their serological results on immunofluorescent tests. Immunofluorescence with Marburg virus was described in BHK21 by Carter and Bright (17) but recent work on Marburg and Ebola has been based on the indirect immunofluorescent technique described by Wulff and Lange (9) for use with Lassa virus. The method is more sensitive than complement fixation and appears to be quite specific. It certainly distinguishes between Marburg and Ebola viruses (2). Wulff and Conrad (18) believe that the method could be used successfully for survey work since, at least with Marburg, antibodies can be detected five years after infection by this method.
The diagnosis of Ebola virus infection was confirmed by the indirect immunofluorescence test in 42 of 48 Maridi patients who had been diagnosed clinically. These results were confirmed in several patients by a parallel test kindly carried out at CDC, Atlanta, by Dr. Patricia Webb. There can be little doubt that the other six patients, in whom no antibodies could be detected, had been infected with the same virus since they all had febrile illnesses accompanied by haemorrhages and symptoms as described by Smith et al. (5). The lack of detectable antibodies in 25 of 31 Nzara patients was surprising and disappointing especially as all of them had had illnesses clinically indistinguishable from proven cases of Ebola fever. The Nzara patients were bled at times ranging from six to 17 weeks after the onset of their illnesses. It may be that their antibody levels may have fallen below those detectable by immunofluorescent testing. In view of earlier results with Marburg virus this possibility seems unlikely. However, antibody levels in proven cases of Ebola infection in Maridi did fall quickly. 18 of 23 patients bled in November 1976 and again in January 1977 had distinct falls in antibody levels ranging from 2- to 5- fold. Later antibody levels were often detected at the lowest acceptable dilution of 1:8. Two patients actually became negative. These results could possibly be explained if the immunofluorescent test was directed towards detecting IgN antibodies alone. Unfortunately the test was specifically directed to detect IgG and therefore no satisfactory explanation for the lack of detectable antibodies can be advanced. Obviously further work needs to be carried out on developing more sensitive test systems.
The finding of Ebola virus antibodies in 25 out of 131 (19%) Maridi case contacts and in seven of 64 hospital staff contacts (19%) indicates that Ebola virus can cause milder illnesses and even subclinical infections. It also supports the view advanced by Francis et al. 6 that Ebola virus is not readily transmissible from person to person except where there is close and prolonged contact with a patient suffering from severe disease. Clinical observation of cases occurring late in the epidemic showed that frank haemorrhagic manifestations were less apparent. There may have been some attenuation of Ebola virus resulting from continuing passage in man. Observation in guinea pigs infected with isolates obtained later in the illness may confirm this.
The antibody results on sera collected from the Nzara cotton factory staff indicate that nine of the 24 staff (37%) of the cloth room and adjacent store were infected. The levels of antibody, ranging from 1:16 to 1:256, indicate a fairly recent infection although the seven members of staff who were bled in November and were shown to have been infected gave no history of illness. These results point very strongly to the cotton factory having been a prime source of infection. The results of virus isolation attempts and serological studies on rodents and bats collected in the factory are awaited with interest.
We are very grateful for the active support of all those people, too numerous to mention individually, in Maridi and Nzara, who helped us in so many ways.
We would also like to thank Dr. Patricia Webb of CDC, Atlanta, for carrying out parallel serological studies on some of the patients' sera.
1. Bowen, E.T.W., Platt, G.S., Lloyd, G., Baskerville, A., Harris, W.J., Vella, E.E. (1977) Viral haemorrhagic fever in southern Sudan and northern Zaire, Lancet, 1, 571-573.
2. Johnson, K.M., Webb, P.A., Lange, J.V., Murphy, F.A. (1977) Isolation and partial characterization of a new virus causing acute haemorrhagic fever in Zaire, Lancet, 1, 569-571.
3. Pattyn, S., Jacob, W., Van der Groen, G., Piot, P., Courteille, G. (1977) Isolation of Marburg-like virus from a case of haemorrhagic fever in Zaire, Lancet, 1. 573-574.
4. Simpson, D.I.H., Francis, D.P., Smith, D.H., Highton, R.B., Pacifico Lolik, Isiaih Mayom Deng, Anthony Lagu Gillo, Oliver Duku, Noel Logo Warille, William Renzi Tembura, Ali Ahmed Idris, Babiker El Tahir (In press) Viral haemorrhagic fever in the Sudan 1976: introduction, Bulletin of the World Health Organization.
5. Smith, D.H., Francis, D.P., Highton, R.B., Simpson, D.I.H., Isiaih Mlayom Deng, Pacifico Lolik, Anthony Lagu Gillo, Babiker El Tahir, Ali Ahmed Idris (In press) Viral haemorrhagic fever in the Sudan, 1976: clinical manifestations, Bulletin of the World Health Organization.
6. Francis, D.P., Smith, D.H., Highton, R.B., Simpson, D.I.H., Pacifico Lolik, Isiaih Mayom Deng, Anthony Lagu Gillo, Ali Ahmed Idris, Babiker El Tahir (In press) Viral haemorrhagic fever in the Sudan, 1976: epidemiological aspects of the disease, Bulletin of the World Health Organization.
7. Ridley, D.S., Simpson, D.I.H., Francis, D.P., Smith, D.H., Highton, R.B., Knobloch, J., Isiaih Mayom Deng, Pacifico Lolik, Babiker El Tahir, Baskerville, A., Bowen, E.T.W. (In press) Viral haemorrhagic fever in the Sudan, 1976: human pathology, Bulletin of the World Health Organization.
8. Simpson, D.I.H. (1977) Marburg and Ebola fever: a guide for their diagnosis, management and control, World Health Organization Offset publication N'36.
9. Wulff, H., Lange, J.V. (1975) Indirect immunofluorescence for the diagnosis of Lassa fever infection, Bulletin of the World Health Organization, 52, 429-436.
10. Ellis, D.S., Simpson, D.I.H., Francis, D.P., Knobloch, J., Bowen, E.T.W., Pacifico Lolik, Isiaih Mayom Deng (In press) The ultrastructure of Ebola virus particles in human liver, Journal of Clinical Pathology.
11. Smith, C.E.G., Simpson, D.I.H., Bowen, E.T.W., Zlotnik, 1. (1967) Fatal human disease from vervet monkeys, Lancet, ii, 1119-1121.
12. Kissling, R.E., Robinson, R.Q., Murphy, F.A., Whitfield, S.G. (1968) Agent of disease contracted from green monkeys, Science, 160, 888-890.
13. Siegert, R., Slenczka, W. (1971) Laboratory diagnosis and Pathogenesis. Marburg Virus Disease: Proceedings of a symposium, Editors Martini, G.A. and Siegert, R., Springer-Verlag, Berlin.
14. Malherbe, H., Strickland-Chomley, M. (1971) Studies on the Marburg virus. Marburg virus disease: Proceedings of a symposium, Editors Martini, G.A. and Siegert, R., Springer-Verlag, Berlin.
15. Slenczka, W., Wolff, G. (1971) Biological properties of the Marburg virus. Marburg virus disease: Proceedings of a symposium, Editors Martini, G.A. and Siegert, R., Springer-Verlag, Berlin.
16. Slenczka, W., Wolff, G., Siegert, R. (1971) A critical study of monkey sera for the presence of antibody against the Marburg virus, American Journal of Epidemiology, 93, 496-505.
17. Carter, G.B., Bright, W.F. (1968) Immunofluorescent study of the vervet monkey disease agent, Lancet, ii, 913-914.
18. Wulff, H., Conrad, J.L. (1977) Marburg virus disease in Comparative Diagnosis of Viral Disease. II. Human and related viruses. Part B. Editors Kurstak, E. and Kurstak, C., Academic Press, London and New York.
A.W. Woodruff : Can I ask our colleagues about their own quarantine ?
D.I.H. Simpson : We have in fact more or less put ourselves in quarantine and had no contact with patients for at least ten days before we left the area, on the 20th of November. The lost case was known to have occurred on the 22nd of November, we didn't have contact with that. We did go into quarantine somewhat enforced and somewhat reluctantly for another fortnight after that.
G.A. Eddy : I am interested in the apparent disappearance of antibodies. This has some importance for epidemiologic studies and also for the collection of immune plasma: . Might it not be possible to have an exchange of sera between your place and Atlanta for instance, so that we can get a little better confirmation of this apparent disappearance of antibodies ?
D.I.H. Simpson : This in fact has already been done in several cases and there is a reasonably good correlation between the results which were carried out at Porton with the ones that were carried out in Atlanta. I should point out of course that all our tests have been carried out with the Zaire 718 strain and it is possible, although I think it is not all that conclusive, that the fact that we are using a Zaire strain to test Sudanese patients my also influence the results somewhat.
J.G. Breman : There are two alarming points in this epidemic, I think both of them have some lessons for us. The first is that Dr. Babiker said the epidemic first came to his attention on the 30th of September. Yet by the 30th, at least as I recall, 75 cases with 3O or 40 deaths had occurred. Can you explain what the delay was there and relate this to possible quicker action in the future ? The second point is the fact that in the Sudan at Least ten generations of cases had occurred before the epidemic died out. In many instances you did a beautiful job in showing five or six generations traced early. Working at epidemics, Looking at monkey parks with a computer, we found that you need a 12.5% secondary attack rate to continue an epidemic without it dying out. With this secondary attack rate on our side so high with close family contacts and on your side actually very high in the first and second generation, this is very alarming compared to, say, Lassa fever rates where it is unusual to have a second and third generation. When you work at the data of only seropositive cases compared to death is there any decay in mortality rates throughout the epidemic which would then account for what happened ? By month there was apparently no difference in mortality.
E.T. Babiker : The outbreak became alarming when it was introduced into Maridi hospital. It became evident to the Minister of Health that there was an alarming situation when there was a large outbreak among the medical staff and when the medical doctor fell ill and was transported to Khartoum. Meanwhile between August and the end of September the medical doctor had a few cases continuously and tried to investigate these, collecting blood for typhoid and sending it to Khartoum, receiving the result or not. In this way it took six or seven weeks to confirm or exclude this infectious agent in such a distant place far from Laboratory facilities. This accounts for the delay. We also asked ourselves if those outbreaks occurred in the t I think this historical information is probably meaningless because the noes are that the outbreak is probably not a new disease. But I think if would not have spread to Maridi and Yambuku hospitals, no one would have remembered it. There has never been any outbreak like these before, because -amplifying forces have not been there : a needle, an amazing remarkable social structure or a newly established teaching hospital where these kind of outbreaks can happen. You need this sort of secondary amplifiers to get a recognizable outbreak and I think even that the 70 cases of the Nzara outbreak, it remained Limited to that locality, would have gone undetected.
Van Nieuwenhove : During the epidemic in Maridi, 13.914 doses of typhoid vaccine were administered. The hospital material for such a Large number of vaccinations nations was certainly insufficient while the sterilization procedures the same as in Yambuku. Could not therefore injections have played an equally important role in the Sudan as in Yambuku ? The cases we had in Sudan have actually been traced back to the I source. There were no cases which could not be explained. Most of the a had had contact with other cases during the incubation period. We saw the same thing in Yambuku, it's not because someone has had contact with a case that necessarily this contact was the source of the infection. In Sudan the death rate for hospitalized patients went up from 25% in August, to 44,6% in September and 70% in October. The vaccination campaign was started on the 25th of September, I find such a rise in death rate very strange.
P. Brès : The vaccination programme was not completed, perhaps only 114 of it was completed. As Dr. Daoud said, man to man transmission was well evidenced and the decrease of cases was a result of the use of protective clothing. When these ran short, there was a booster effect, and this illustrates the efficiency of the protective clothing. As you know, most of the victims were the not weIl-trained student nurses in the hospital and these were eliminated from the second phase. it is dangerous to stick too much to the figures if you don't know what was behind them.
D.P. Francis : We did out very best to administer the typhoid and gamma globulin injections that were brought in early before the diagnosis was known. But actually very few doses were given, a lot of them in the hospital staff because they were at high risk and could obtain the necessary needles and syringes. Since they had some knowledge they did not shape this material with any active cases on the ward.
A.W. Woodruff : Among 13.500 injections given, under the circumstances one would expect that afterwards there would be a fair incidence of hepatitis B infection.
E.T. Babiker : This number of vaccinations is absolutely incorrect. We have not been thinking very much in terms of injection transmission in the Maridi outbreak. I don't think the magnitude of the outbreak in Maridi is compatible with injections. it could be for a few field cases but I think the main transmission was by person to person transmission. As to the hepatitis cases, we have not any.
J. De Smijter 1 do not agree that one would see many cases of hepatitis after even when it is massive. In a survey in Mali we found evidence of past hepatitis B in 80% of the population. In Zaire we found 60 to 70%, this means t hat more than 90% of the population has immunity against hepatitis B. In that case I would not expect to see many cases of hepatitis B after vaccination.
D.P. Francis : As has been said, the immunization programme in Maridi was very limited. Contrary to the situation in Yambuku, there was no prenatal clinic nor a well baby clinic to serve as an amplifier. I think there was no needle contact between well and ill people. There were people infected in the hospital, admitted initially for various diseases, who were treated in the hospital, including injections, and who five to seven days later came down with usually a fatal disease. There are six definite Ebola cases nosocomially acquired which I presume were needletransmitted. ,-,'a do not have any other documentation of people who came to the hospital prior to their disease and who received injections. Some of these no doubt aid occur in Nzara where nurses did private practice carrying needles around. The worst people in washing their hands between patients ape physicians. In a country that has a dollar per patient per year to spend on medicine, the thought of providing plastic gloves for every febrile patient is just not possible.
M. Isaäcson : Maridi hospital was a large teaching hospital and it was noted that the main victims were the student nurses. In teaching hospitals, through out the world student nurses are in the unfortunate position of frequently having to do the most unpleasant duties and these are the people who are least trained and least experienced to deal with this safely. In future nurses training there should be a great deal of emphasis on safety precautions. Particularly in Africa any pyrexia of unknown origin should be regarded potentially a lethal viral haemorrhagic fever and should be treated and handled as such. In addition, when there is any suspicion of a lethal haemorrhagic fever being dealt with, one of the most important approaches is to have constant supervision and guidance from someone who is skilled in the protection of hospital staff and in the protection against hospital cross infection.
R.A. Coutinho : I can understand that people panicked since they realized that the treatment which was given in the hospital was not very successful. As far as I can say from my experience in Africa, people would then go to the witch doctor. I didn't hear that, but I could imagine that the witch doctors were giving some treatment and possibly that they played a role. I wonder if anyone inquired about this.
J.G. Breman : There was one cluster of nine cases in Zaire which was interesting in that eight of these were women who received scarifications all over the body following the death of a child with whom another was in contact. All of these women died about the same time, but they did have contact as well as this very interesting practice. Another practice which I felt was even more important was a type of purgative which is used throughout Africa for a variety of treatments, but here it was done in a very interesting way. In this little group it was done with a thin bamboo reet put into the rectum and then the others would blow material through this reet. The same reet was used for all the women. Of course this could have incised the intestinal mucosa and certainly implicates blood products or stool as a vector.