CLINICAL ASPECTS OF EBOLA VIRUS DISEASE AT THE NGALIEMA HOSPITAL, KINSHASA, ZAIRE, 1976
MARGARETHA ISAACSON(1), P. SUREAU(2), G. COURTEILLE(3), S.R. PATTYN(4)
1. The South African Institute for Medical Research, Department of Epidemiology, P.O.Box 1038, Johannesburg, South Africa 2000.
2. Pasteur Institute, Teheran, Iran.
3. Ngaliema Hospital, Kinshasa, Zaire.
4. Institute of Tropical Medicine, Antwerp, Belgium.
The epidemic of Ebola Virus Disease (EVD) which had been in progress in Yambuku in northern Zaire for some time had already taken a heavy toll in lives among the staff at the Yambuku Mission Hospital when the first Belgian nun, a midwife, became ill on September 14th and died on the 19th. The second of the three Belgian nurses, Sister M.E. (Kinshasa Case 1), aged 42 years, became ill 4 days later and she was transferred to Kinshasa as lack of staff caused the hospital to close. Accordingly, she left Yambuku in the company of a nursing colleague, Sister E.R., and a priest, Father A.S. They travelled by road to Bumba where they spent the night at the local convent and the next day flew by scheduled flight of Air Zaire to Njili, Zaire's International Airport near Kinshasa, from where they took a taxi to the Ngaliema hospital where she was admitted on 25 September. Sister E.R. continued to nurse Sister M.E. who died September 30th. Sister E.R. (Kinshasa Case 2), aged 56 years, became ill with similar symptoms 8 days later and died on October 14th. Two days earlier Zaire nurse, M.N. (Kinshasa Case 3) aged 23 years, employed at the Ngaliema Hospiral who had nursed Sister M.E. also became ill (Fig. 1).
Examination of liver tissue taken post-mortem from Sister M.E. resulted in diagnosis of Marburg-like disease at the Institute for Tropical Medicine in Antwerp, Belgium, and a request was issued by the Zaire Government for specific Marburg convalescent plasma.
A supply, accompanied by a physician, arrived from Johannesburg, South Africa, on 16 October. By then it had been established at the Center for Disease Control Atlanta, Georgia, USA, that the virus, now known as Ebola virus, although morphologically similar to Marburg virus, was antigenically different. It was therefore believed unlikely that the Marburg plasma would be of benefit but two units (500 ml) were nevertheless administered intravenously to Nurse M.N. on on October 16th. However, her condition continued to deteriorate and she died on October 20th.
Fig. 1 Interrelationships between 3 fatal cases of Ebola Virus Disease in Kinshasa, Zaire, 1976.
Clinical features. These are summarized in fig. 2. The early symptoms include fever, headache, myalgia, diarrhoea and vomiting, which are non-specific and may not indicate the serious and highly lethal nature of the infection. The characteristic triad of features which leave little doubt about the diagnosis, namely haemorrhage, rash and severe sore throat occur later during the course of the infection at a stage when its progression to death may be irreversible.
The rash in these three cases was morbilliform and started on the front of the trunk on day 5 or 6, spread to the back, buttocks and limbs on the following day and disappeared the day after.
On examination the throat of Case 1 showed reddening in the early stages which progressed over the next few days to severe, red, oedematous, tender swelling of the soft tissues which caused great difficulty in swallowing.
In Sister M.E., severe dysphagia as well as dyspnoea resulted from the swollen tissues at the back of the throat which was preceded by intense reddening of the tongue and throat on day 3 when she also had conjunctival injection.
Haemorrhage was manifested in Case 1 by oral and conjunctival petechiae on day 4 of illness, haematemesis and melaena from day 5, gingival bleeding on day 7, and bleeding injection sites on day 8. Case 2 had melaena from day 6. Case 3 had one slight haematemesis with fresh blood on day 7, susbsequent vomitus being free of obvious fresh or altered blood. On day 8 she had some melaena, and large echymoses developed, especially over pressure points such as elbows and shoulders. She also manifested marked swelling of the face and upper extremities on day 7 of illness. Urinary output at this stage was good and in balance with fluid intake.
Case 1 exhibited erythematous swelling of the vulva. All three patients were mentally alert until shortly before death. Anxiety was very marked in Case 3.
Laboratory findings. Laboratory tests were done in Case 1 (fig. 2) but, to safeguard the health of laboratory personnel, no tests were done by the hospital laboratory on the other two patients. However, the International mission (IMC) established some basic laboratory services within the confines of Pavillion 5 and some tests could therefore be done on Case 3 in the later stages of illness.
In none of the three patients were blood counts done sufficiently early to establish whether a marked leukopenia occurs in EVD as it does in disease (MVD)(3,4,5). Platelet counts were unexpectedly normal, unlike those found in MVD.
In case 3, mild clinical jaundice was diagnosed on the basis of yellowed sclerae on the day before death. Although not noted in the clinical record, jaundice may also have occurred in the terminal stages of Case I where a total serum bilirubin of 59,8 micromol/l (3,5 mg/100 ml) was established. Raised SGOT and SGPT levels which were established in only one patient indicated the occurrence of liver damage.
The only laboratory evidence of the probable occurrence of disseminated intravascular coagulation (DIC) was obtained from Case 3 in whom tests for fibrinogen degradation products were done which were shown to be present in increasing amounts.
Ebola virus was isolated from blood of all three patients i.e. from Case 1 on day 6, from Case 2 on days 3 and 6, and from Case 3 on day 3. In addition, formalinized liver tissue showed, by electron microscopy, the presence of virus 2 particles resembling those of the Marburg group of viruses.
Fig. 2 Clinical and laboratory data of three fatal cases of Ebola Virus Disease in Kinshasa, Zaire, 1976.
Case 1. When Sister M.E. was admitted to hospital, the aetiological agent responsible for the epidemic was still undetermined and typhoid was high on the list of differential diagnoses. Sister M.E. was therefore put on cotrimoxazole which was later changed to chloramphenicol and penicillin. Vitamin KI was given daily from day 4 onwards and she received blood transfusions on days 5, 7 and 8. Intravenous fluid therapy was started on day 5. in spite of the regression of some symptoms, i.e. dysphagia and vomiting, her condition progressively deteriorated and in spite of supportive treatment in the form of adrenalin and hydrocortisone on day 8 she died shortly thereafter.
Case 2. In view of the increasing suspicion that a viral agent was responsible for the epidemic, an antiviral preparation, 'Virustat' was given to this patient. She, like Case 1, was given aspirin as an antipyretic. Gamma globulin was administered on days 3 and 5, and chloramphenicol was started on day 3. Enterovioform was used in an effort to control the diarrhoea which was a major feature in this case. Intravenous fluid replacement was started on day 6. The patient's condition deteriorated; she was given hydrocortisone on day 6 and she died on day 7.
Case 3. The first three days of this patient's illness were marked by nothing more specific than pyrexia and fatigue. A provisional but unconfirmed diagnosis of malaria was made at one of the hospitals she visited and she was eventually admitted to her own ward at the Ngaliema hospital where antimalarial therapy was continued. On day 4 her oral fluid intake dropped to negligible levels necessitating intravenous therapy. When she was admitted to hospital it became known that the cause of the epidemic was a Marburg-like virus. Consequently on day 4 she was given 2 units (500 ml) of Marburg convalescent plasma. Valium was given to relieve her anxiety and chloramphenicol was also administered.
The severe sore throat was greatly relieved by the sucking of ice cubes. In view of the favourable outcome gained with two Marburg patients treated with anticoagulation(5) heparin treatment was started on day 6 in anticipation of DIC when the patient had developed all the classical symptoms of EVD but clinical haemorrhage had not yet become evident. Heparin was prescribed as a continuous infusion at the rate of 16 000 U daily on day 6 (with a loading dose of 2 000 U) increased subsequently to 30 000 U/24 hours. There were problems with the intravenous apparatus and for prolonged periods of time, intravenous therapy was intermittent, resulting in unsatisfactory anticoagulation as evidenced by the PTT which had not significantly increased. Although haemorrhagic phenomena developed these were not severe in terms of actual blood loss.
On day 7, the pulse rate increased sharply; the patient complained of substernal pain and developed a gallop rhythm with a pulse rate of 136 beats/minute. She was digitalized which resulted in a slowing of the heart-rate but she died that night, possibly as the result of a diffuse myocarditis.
The clinical features of these three cases of EVD are virtually indistinguishable from those seen in MVD. The high case fatality rate might be reduced in intensive care facilities where the necessary expertise and equipment are available for handling cases with the problems associated with extensive disseminated intravascular coagulation.
Anticoagulation is still a controversial subject as far as its benefits in this kind of case is concerned. In South Africa two out of three MVD patients with laboratory and/or clinical evidence of DIC were given very carefully regulated and monitored prophylactic heparin treatment, and both recovered(5). The full-blown case with severe DIC who is already depleted of clotting factors should not be anticoagulated but given replacement therapy in the form of fresh frozen plasma.
Although the white cell counts were found to be normal, none were done at the onset of illness and the leukopenic phase was probably missed.
One of our three patients had an elevated serum bilirubin level in the terminal stages and slight jaundice of the sclerae was noted in a second case.
It should be stressed, however, that clinical jaundice is not a common observation and, when present, is very slight in spite of the profound hepatic damage which may occur.
Although it was shown in vitro that the Ebola virus was antigenically different from the Marburg virus this did not necessarily imply a lack of crossimmunogenicity. For this reason Case 3 was given the benefit of the doubt, and plasma containing Marburg antibody to a titre of 1:64, obtained from a South African nurse, was administered, but the patient, not unexpectedly, failed to respond.
A hospital-based outbreak of Ebola Virus Disease (EVD) occurred in Kinshasa, capital of Zaire, following the arrival of a sick nursing sister from the northern epidemic area. Her travelling companion, and a Kinshasa nurse who cared for her after her arrival in Kinshasa, also became ill and all three patients died after illness lasting 7 to 8 days. The illness was characterized by fever, headache, myalgia, diarrhoea, vomiting and later on a morbilliform rash, sore throat and haemorrhagic phenomena.
The clinical picture closely resembled that of Marburg virus disease (MVD) but the causative agent which is morphologically indistinguishable from the Marburg virus, is antigenically different. It was not unexpected, therefore, that Marburg convalescent plasma, administered to the last Kinshasa case, did not result in a favourable outcome of the illness.
1. Pattyn, S. et al. (1977) Isolation of Marburg-like virus from a case of haemorrhagic fever in Zaire. Lancet, i, 573-574.
2. Johnson, K.M., et al. (1977) Isolation and partial characterization of a new virus causing acute haemorrhagic fever in Zaire. Lancet, i,569571.
3. Martini, G.A. (1971) Marburg Virus Disease. Clinical Syndrome, in Marburg Virus Disease, ed., Martini, G.A. and Siegert, R. New York Springer,pp. 1-9.
4. Stille, W., Böhle, E. (1971) Clinical Course and Prognosis of Marburg Virus ('Green-Monkey') Disease, in Marburg Virus Disease, ed., Martini,G.A. and Siegert, R. New York, Springer, pp. 10-18.
5. Gear, J.S.S. et al. (1975) Outbreak of Marburg Virus Disease in Johannesburg. Brit. Med. J., 4, 489-493.