Previous PageIndexNext Page


1. The South African Institute for Medical Research, Department of Epidemiology, P.O.Box 1038, Johannesburg, South Africa 2000.
2. National Institute for Virology, Johannesburg, South Africa.
3. Center for Disease Control, Atlanta, U.S.A.
4. French Army Medical Services, Marseilles, France.


The International Medical Commission (IMC) which was formed by the Minister of Health of Zaire initially consisted of about a dozen members from various parts of the world. Subsequent recruiting in Zaire and elsewhere to establish the many teams required for field surveillance and other duties brought the total to approximately 50 members, mostly of the medical, nursing and paramedical professions but also administrative personnel and mechanics. The latter were required for maintenance of the Commission's fleet of vehicles, radio communication, power supply installation, equipment stores, etc.

The IMC's function was to investigate the cause, clinical manifestations and epidemiology of Ebola Virus Disease (EVD) and to advise and assist the Ministry on measures of control.

Exposure to infection by its members in the field, in the hospitals and in IMC laboratories was expected to occur and, at the time of the IMC's inception, there was every reason to believe that at least some members would develop clinical disease. It was necessary to establish contingency plans to ensure that optimum medical care facilities would be made available. The Yambuku mission hospital had ceased functioning as a health care centre due to the death of almost all of its personnel and the IMC had to provide some of the most basic emergency services such as minor surgery and obstetrics.

An improvised intensive care isolation unit was established to receive suspect EVD cases from the local population but it was recognized by the IMC that it would not be possible to give patients all the benefits of modern medical science under the locally prevailing conditions. The hospital is situated in the Zaire River basin where the vegetation consists of dense equatorial rain forest cover. Roads are non-existant and only a dirt-track requiring 4-wheel drive vehicles gives access to Yambuku, epicentre of the outbreak. All supplies had to be airfreighted by military aircraft, and helicopters were made available to the IMC as commercial air services had ceased its scheduled flights to the epidemic area which had been placed under strict quarantine.

From time to time the generators providing electric power broke down and could have caused critical interruptions of patient monitoring and therapeutic equipment.

Furthermore, the need to care for critically ill friends and colleagues under prevailing sub-optimal conditions would impose a serious additional workload and could be expected to have adverse psychological effects.

Taking cognizance of all these uncertainties and difficulties the IMC decided against nursing its members locally, even with the assistance of fully equipped intensive care teams which could be brought in from elsewhere. It was therefore agreed that IMC members suspected of having contracted EVD would be evacuated.

As evacuation could have serious financial and other repercussions a strict surveillance of IMC members was instituted from the start and criteria for evacuation were established.

Medical surveillance of IMC members. All IMC members were issued with personal medical record forms as reproduced in Fig. 1 and individual thermometers were supplied. Malaria, being hyperendemic in the area would be high on the list of differential diagnosis in cases with PUO (pyrexia of undetermined origin) and adequate prophylactic treatment was therefore prescribed. Only boiled water and bottled beverages were used for drinking purposes and meals were cooked at the mission in order to minimize the incidence of typhoid and other intestinal infections. Typhoid in particular poses problems in the differential diagnosis of EVD.

IMC members were also expected to record the dates on which actual or potential exposure to Ebola virus occurred. This information would be required when deciding whether a PUO was likely to be due to Ebola virus or to some other agent.

Criteria for Medical Evacuation of patients suspected of having EVD.

While there was clearly a need for evacuating patients without undue delay it was equally important for patients to be carefully evaluated in order to avoid unnecessary implementation of the evacuation procedures which would be very costly and require international team work.

Major criteria, all of which had to be fulfilled, were

1. Pyrexial illness with one or more other symptoms of EVD of 48 hours duration.

2. A definite history of actual or potential exposure to Ebola virus within 21 days prior to onset of illness.

3. No other demonstrable cause.

4. Failure to respond to anti-malaria treatment.

Evacuation of such patients was to be implemented 48 hours after onset of illness.

Selection of the receiving patient-care centre. Three criteria were used in the selection of the hospital which was to receive IMC members with suspected EVD.

1. Transit time, as determined by distance.

2. Agreement by the hospital and the national health authorities of the country concerned to accept any number of IMC members for isolation and treatment.

3. Prior experience in treating cases of this kind .

Final arrangements were made with the South African Government as Johannesburg is at a reasonably short distance from Kinshasa and travelling time, in theory, should be minimal. The South African Government offered all possible assistance when advised of the position.

Means of transport.

Both minimum and maximum transit times were calculated for the journey from Yambuku to Johannesburg (fig.2).


Yambuku - Bumba by military helicopter

1/4 hour

Bumba - Kinshasa by chartered commercial airliner

3 hours

Kinshasa - Johannesburg by chartered commercial airliner

3 1/2 hours

Inclusive of transfer between the above means of transport it was expected that 8 hours would be the minimum delay between departure from Yambuku and arrival at the Johannesburg Isolation Hospital.


Yambuku Bumba by Landrover

4 1/2 hours

Overnight delay (airstrip at Bumba not equipped for night flying)

12 hours

Bumba Kinshasa by chartered commercial airliner

3 hours

Delay for USAAF aircraft to arrive in Kinshasa from Europe

8 hours

Kinshasa Johannesburg by air

3 1/2 hours

Inclusive of transfers the maximum delay was calculated at

32-34 hours.

It was our opinion that transport of a patient should not be undertaken beyond the quarantined epidemic region unless some form of isolation could be provided to prevent spread. Arrangements had therefore been made to obtain an aircraft negative pressure transit isolator. This was designed to fit into most aircraft used by commercial airlines. An alternate arrangement was the use of a US NASA isolation capsule. Its size necessitated also the provision of an aircraft by the USAAF large enough to accommodate the capsule as no aircraft currently in use by commercial aircraft could load this on board through existing access ports.

In the absence of either means of patient isolation it was envisaged that low flying, thus eliminating the need for pressurization, could safeguard aircrew from droplet infection.

Contingency plans were made accordingly by the three governments concerned in anticipation of EVD occurring among IMC members.

Implementation of contingency plans. On the evening of Friday 26 November a male US Corps volunteer aged 27 years complained of fatigue, cold shivers, headache and backache. His temperature was 37,1ºC and it was decided to review the situation the next morning. The following day his temperature had risen to 37,6ºC and he complained of abdominal pain, nausea, mild diarrhoea and urinary frequency. A full medical examination, using protective clothing, was carried out but nothing of significance was found. Microscopic examination of a urine sample showed no abnormalities, the WBC was 4680 with normal differential count, platelet count was 143 000 and no blood parasites were detected.

The clinical picture was compatible with the early stages of EVD, but also with that of some other infections. The patient was put on therapeutic doses of chloroquin and cotrimoxazole.

The patient's Ebola-virus exposure record was reviewed. He had been engaged in taking blood samples in the villages but none of the people he had bled had been shown to be active or convalescent cases of EVD. He had also assisted in the virology laboratory in Yambuku where he had handled and microscopically examined, but not prepared, slides with ultraviolet irradiated Ebola virus. It was shown that a recently arrived batch of this antigen still contained live virus in spite of UV treatment. The patient therefore had a potentially positive recent exposure history.

Fig. 2 Map of Africa showing the various localities involved in the 1976 outbreak of Ebola virus disease in Sudan and Zaire. Medical evacuation of an IMC member suspected of having EVD took place from Yambuku to Johannesburg via Bumba and Kinshasa.

The IMC logistics base in Kinshasa was alerted to the possibility of EVD in a staff member and was requested to alert others concerned without however implementing evacuation procedures. In accordance with the contingency plans it was decided to wait a further 36 hours in order to exclude other conditions for which the patient was being treated. That evening his temperature reached 38,3ºC with a relative bradycardia of 84 beats/minute. Blood pressure was normal. He was isolated in his room and lock-up toilet and shower facilities were made available for his exclusive use. No visitors were allowed and all medical and nursing care were supplied by two IMC physicians.

On 28 November the patient remained mildly pyrexial and his gastro-intestinal symptoms and urinary frequency had ceased. His WBC had decreased to 4250 with a marked absolute and relative increase of monocytes to 24% of the total WBC.

The patient was moved to the improvised isolation intensive care unit in the hospital. He had proteinuria, a constant finding in all the EVD patients. The logistics base was requested to implement the evacuation programme to the point of full stand-by and readiness for take-off. The aircraft transit isolator had just arrived in Kinshasa. As the patient seemed clinically somewhat improved and the 48-hour period would terminate at nightfall the final decision whether to evacuate the patient was deferred till 6 am the following morning. That morning (29 November) the patient was still pyrexial and he had developed severe generalized backache. His back muscles were tender to palpation, he was nauseous and his pharynx was injected. The IMC logistics vase was requested to inform the governments concerned that evacuation would take place with immediate effect. By implication this was an indication for the administration of convalescent serum. Two units were heated in a 58ºC waterbath for 1 hour to inactivate residual Ebola virus. It was then centrifuged and re-transferred to a new bag to eliminate large coagulated masses which had formed during heating. A power failure occurred during centrifugation. This also interrupted radio communications and departure was of necessity delayed.

Contrary to expectations a helicopter was not available and a landrover was equipped to install the patient as comfortably as possible. He was dressed in disposable surgical clothes and accompanied by the two attending physicians one of whom acted as driver. Both wore disposable protective clothing including masks. A spare landrover driven by another IMC member was stocked with drugs and equipment to deal with medical emergencies. The forest tract was in poor shape and the patient in severe pain in spite of analgesia. The drive to Bumba therefore took all of the anticipated 4 1/2 hours. Temperature of the patient was 38,7ºC. It had been decided to administer the convalescent plasma after boarding the aircraft in Bumba where conditions would be more comfortable. The aircraft was scheduled to have left Kinshasa early in the morning and to arrive in Bumba before the patient. This was not the case and air transport did not arrive until the following morning. The landrover with the patient remained on the airstrip for some time while the IMC member in the spare landrover left to enquire about the fate of the aircraft. Such was the fear of the local populace that parking facilities in the shade of the control tower were denied to the IMC party. To protect the public the patient wore a half-face respirator with filter cartridges. He was sedated with analgesics and Valium and an intravenous catheter was inserted in the left arm. A catheter was used to ensure a troublefree intravenous pathway until arrival in Johannesburg in case of potential medical emergencies occurring during the rest of the journey which was already becoming protracted.

The first unit of convalescent plasma was infused in the landrover, very slowly at first and with syringes containing adrenalin and solucortef at hand. This was followed by slow infusion of 5% dextrose saline. The second unit of plasma was administered that evening after accommodation was obtained in the Bumba guest house with another IMC member.

An Air Zaire 'Fokker Friendship' arrived at dawn together with the aircraft transit isolator (Fig. 3). Appropriate decontamination of linen and utensils used by the patient was carried out at the guest house. He was still pyrexial and complained of severe backache and nausea. Wearing a clean disposable full length gown and half-face respirator the patient was transferred into the isolator without touching its external surfaces. The isolator was sealed after the air extraction mechanism was switched on. This was a prototype apparatus and difficulty was experienced in anchoring the bottle of intravenous fluid in an elevated position. An improvised anchor was devised. During take-off, and subsequent landings and take-offs, constant attention was required to the intravenous apparatus as the infusion rate fluctuated markedly. The sides of the plastic canopy tended to draw in unduly due to excessive suction which we did not know how to regulate. These faults apart, the patient was very comfortable and no problems were experienced in attending to his needs. Kinshasa was reached at 13h00 and a further disappointment awaited the evacuation team when it was learnt that no local aircraft was available for the next leg to Johannesburg. A USAAF C141 'Starlifter' had therefore already left Madrid for Kinshasa where it was expected to arrive that evening at 1900 hours. The next six hours were spent by the patient in the isolator on the back of a truck parked in a remote hangar. It was excessively hot and humid, the isolator started fogging up, the patient's temperature rose to 39ºC and he was in a great deal of pain and discomfort. He was heavily dosed with analgesics and Valium and a battery operated fan was procured. His stock of bottled mineral water was used for a sponge-down in an attempt to prevent a further rise in temperature. These measures were successful in making his condition much more comfortable. During his period in the hangar he started oozing blood from the intravenous puncture site. The USAAF transport aircraft arrived on schedule and took off for Johannesburg at 20h30 by which time the patient was drowsy. Turbulence was expected on the direct route and it was feared that this might precipitate vomiting and problems with the intravenous apparatus. The latter had become important in view of the possibility of bleeding becoming more severe and requiring supportive measures.

A detour was therefore made over the Atlantic Ocean, prolonging the flight by several hours but resulting in a very smooth journey. At arrival in Pretoria the isolator transport to the Johannesburg isolation hospital was handled by the South African Air Force. On arrival at the hospital, i.e. early on day 6 of illness the patient was noted to have a florid morbilliform rash which lasted two days. His platelet count had dropped to 105 000 and his WBC to 4000. His temperature returned to normal on day 6 and he made an uneventful recovery, nursed in the negative pressure containment bed isolator. Details of handling this patient in the bed isolator have been reported elsewhere (1,2). During convalescence his WBC rose to a normal value of 7700 and platelets to 300 000. Other clotting factors as well as liver function tests gave normal values from day 6 onwards. Specimens of urine, blood and throatswabs taken daily from day 2 of illness failed to yield virus, nor could antibody to known viruses or other pathogens be demonstrated.

Fig. 3. The Air Zaire `Fokker Friendship', one of the IMC's landrovers with the patient, and the negative pressure isolator parked on the Bumba airstrip. The two escorting physicians are shown, one attending to the patient in full protective clothing. The isolator has been placed with its entry port facing the Landrover in readiness for the patient's transfer.

A great deal was learnt from this experience. The following are worth noting:

1. The problems posed by a single patient with a PUO and an uncertain exposure history to a lethal virus. The mild oozing of blood on day 5 and the appear ance of the characteristic morbilliform rash on day 6 left no doubt in our minds at this stage that this was a case of Ebola virus infection. Subsequent negative laboratory findings were unexpected. Exhaustive investigations in the USA and in South Africa have so far failed to identify the etiological agent in this case. The possibility that the disease was caused by an as yet unknown agent cannot be excluded. A confusing factor was presented by a history of a bite by a half-wild pet civet cat just prior to the patient having joined the IMC, i.e. about 2 weeks before onset of illness. Exposure to rabies could therefore not be excluded but as the patient had been given a full course of duck embryo vaccine (a routine vaccination for Peace Corps volunteers) rabies was not considered a likely diagnosis. It is possible however that the civet cat was the source of another pathogenic agent which may have been the cause of the patient's illness. It is rather curious that a civet cat also featured in the history of the index case of the 1975 South African Marburg virus disease outbreak, although no aetiological association could be established. Treatment was continued with human diploid cell vaccine. Similar problems in deciding whether a patient with PUO from endemic or potentially endemic areas of Lassa fever, Marburg virus disease etc., should be dealt with as an international public health hazard are repeatedly experienced in various parts of the world (3,4,5).

2. The usefulness of a transport isolation system. The aircraft negative pressure transit isolator prototype was tested under the worst possible circumstances and found to be easy to handle. The need for several improvements or modifications were communicated to the manufacturers and these have since been incorporated in subsequent models. No problems we encountered at any stage with the power supply. Spare batteries were available and used while used set was recharged during our delay in Kinshasa.

3. Although a transport isolator was included in the contingency plans, its availability was uncertain at the time the patient became ill. It arrived unexpectedly and had to be assembled in Kinshasa in somewhat of a hurry. Likewise, the South African authorities were faced with the welcome but un expected arrival of a containment bed isolator, on loan from the CDC, about 12 hours before arrival of the patient. No assembly or operating instructions accompanied the consignment, but the concerted efforts of several organiza tions succeeded in having it operational by the time the patient arrived. The isolator was designed to operated on 110 Volt, 60 Hz current but local current is 220-250 Volt, 50 Hz. An adjustment was therefore made by means of a large rhestat, but as the cycles could not be changed this resulted in noisy operation which was troublesome to the patient, Of necessity it should be anticipated that international co-operation may repeatedly result in the movement of bed isolators to areas with different power supplies. Our experience showed that this need not be a handicap since, apart from low-level noise, the unit operated without any problems.

4. As a result of the uncertainty about the means of isolation to be used during the patient's transport the South Africans were faced with the need to prepare for the arrival of the NASA space capsule, weighing approximately 3 tons and of large dimensions. An 80 ton low-loader tank transporter was made available by the S.A. Defence Force for the transport of. the capsule to the hospital site. Facilities were also at hand for the transport of smaller isolators with unknown voltage requirements.

5. Although the landrovers and isolator were stocked with every available item and drug that might be needed during an emergency, a file for opening vials had been forgotten and its lack caused some uncomfortable moments when this became evident in Bumba. It is therefore strongly recommended that detailed checklists of equipment and drugs are prepared in advance and kept with the transit isolator. It is also highly advisable to stock up the isolator itself with all routine necessities prior to receipt of the patient as it is time consuming to pass materials into the isolator once the patient is sealed inside.


Contingency plans were drawn up in the event of an IMC member in Zaire becoming ill with suspected Ebola virus disease (EVD). These provided for the following:

1. Continuous medical surveillance of IMC members.

2. Criteria for medical evacuation of IMC members suspected of having AHF.

3. Selection of the receiving patient-care center.

4. Means of transport, and isolation of the patient during transport.

The contingency plans were implemented when an IMC member developed a PUO conforming to all our criteria for medical evacuation. Cognizance was taken of the need for avoiding undue delay in getting a patient to an appropriate treatment facility while taking care to evaluate sick staff members as thoroughly as possible in order to avoid unnecessary evacuations.

Unanticipated problems occurred and prolonged the transport time beyond the 34 hours calculated maximum. A negative pressure prototype aircraft transit isolator was used and, apart from minor design faults, this was found to be extremely useful and easy to operate. No major problems were encountered in the administration of nursing and medical care through the plastic module which was put to the test under the most adverse conditions in the heat and humidity of equatorial Africa. A number of minor modifications recommended to improve overall efficiency and patient comfort have been incorporated by the manufacturers in subsequent models.

1. Clausen, L., Bothwell, T.H. et al. (1978) An approach to the hospital admission and clinical isolation of patients with dangerous infectious fevers. S.Afr. Med. J., 53, 238-242.
2. Gomperts, E.D., Isaäcson,M. et al. (1978) An approach to the handling of highly infectious material in a routine clinical pathology laboratory and in a viral diagnostic unit. S.Afr. Med. J., 53, 243-248.
3. Editorial (1977) Ebola Virus Infection. Brit. Med. J., 2, 539-540.
4. WHO (1976) Suspect case of Lassa Fever. Weekly Epidemiological Record, 33, 264.
5. CDC (1976) Possible Lassa Fever - Washington DC. Morbid. Mortal. Wkly. Rep., 25(8),64.

Previous PageTop Of PageNext Page