Lessons learnt from the management of a case of Lassa fever and follow-up of nosocomial primary contacts in Nigeria during Ebola virus disease outbreak in West Africa

Study area

The study was conducted at the University of Nigeria Teaching Hospital (UNTH), Ituku/Ozalla, Enugu in Enugu State, south-east Nigeria. UNTH is the largest tertiary/referral hospital in south-east Nigeria and receives referrals from other neighbouring states in south-east, south–south and north central Nigeria. Enugu State has a population of about 3.3 million 19. Lassa fever is not endemic in Enugu State, but outbreaks of the disease have been reported in Imo, Anambra, Ebonyi and Rivers States all of which are within 75–240 km of Enugu. In August 2014, a primary contact of the index case of Ebola in Nigeria visited Enugu during which she took ill but did not spread the disease before she was evacuated back to Lagos State for diagnosis and treatment.

Ethical approval was obtained from the Health Research Ethics Committee of the UNTH.

Data collection

The clinical/epidemiological details of the patient (subsequently referred to as the index case) were obtained from the case note, while information on laboratory diagnosis and further care up to the time of his death were obtained from the Institute of Lassa Fever Research and Control (ILFRC), Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria.

Data on contact tracing and surveillance were prospectively collected. On 12 October 2014, an infection control/surveillance team (ICST) was mandated by UNTH management to undertake contact tracing, risk assessment, possible PEP and decontamination of the environment. The ICST comprised of two infectious diseases’ physicians, a virologist, a public health physician, a surgeon, two infection control nurses, a pharmacist and a laboratory scientist. A phone-based alert management system was established for easy access to the ICST. A list was compiled which comprised of all individuals who were in the vicinity where the index case was managed. Exposed persons outside UNTH (both in the referring hospital and the community) were investigated. Any person ill within 21 days of proven or suspected physical contact with the index case, his body fluid or laboratory samples qualified as a primary contact.

Between 12 and 13 October 2014, a member of the ICST held physical interview with any available primary contact, while phone interviews were conducted for those who were not physically available to assess risk of exposure. Information obtained during the interviews was registered in an alert form and further verified with ward supervisors where possible. The contacts were assigned to one of three categories of risk depending on their level of exposure as follows:20

No risk: No direct contact with the patient or body fluids,

Low risk: Direct contact with the patient (e.g. routine medical/nursing care, handling of clinical/laboratory specimens) while wearing personal protective equipment appropriately,

High-risk: Unprotected (or inappropriately protected) exposure of skin or mucous membranes (e.g. mucosal exposure to splashes, needle stick injury) to potentially infectious blood or body fluids, including unprotected handling of clinical/laboratory specimens.

Active monitoring began on 13 October 2014 for all primary contacts. Self-measurement of axillary temperature was performed by the contacts every morning and evening until 21 days counting from the last day of exposure to the index case. Anyone who reported fever (axillary temperature ≥ 38⁰C) was further evaluated by the infectious diseases’ physician. All high-risk cases were exempted from work for an average of 14 days.

Laboratory investigations

Laboratory confirmation of LF was based on a positive test using Lassa virus-specific reverse-transcriptase polymerase chain reaction (RT-PCR) as previously described by Asogun et al [21]. The primary contacts were subjected to laboratory investigations including complete blood count, urinalysis, coagulation studies, liver enzymes and serum urea/creatinine in order to assess for ‘surrogate markers’ which were defined as abnormalities known to commonly occur in VHFs. Blood film microscopy for malaria parasite was performed for the index case and primary contacts that developed fever.

Ribavirin post-exposure prophylaxis

Post-exposure prophylaxis was offered to consenting primary contacts who had high-risk exposure using oral ribavirin (32 mg/kg stat, then 10 mg/kg 8 hourly for 10 days). Written informed consent was obtained before drug prescription after the benefits, and possible side effects of the drug were explained to them. Adherence counselling was offered. Administration of the drug with food was recommended to limit the side effects. Information on adherence as well as side effects of ribavirin was obtained. Pregnancy, breastfeeding, known hypersensitivity to ribavirin and lack of consent were contraindications for its use.

Statistical analysis

The Epi Info statistical software version 3.5.3 (CDC, Atlanta, GA, USA) was used for analysis. Data were presented as proportions for discrete variables and means ± standard deviation (SD) for continuous variables. The chi-square test or Fisher's exact test as appropriate was used to compare proportions. P values <0.05 were considered significant.

Case description

On 9 October 2014, a 28-year-old man presented to the Emergency department of UNTH having been referred from a private hospital in Port Harcourt, Rivers State, with a 10-day history of fever, urinary and chest symptoms, and a diagnosis of sepsis. He had no precise exposure history for VHF. Except for axillary temperature of 39.3⁰C, his physical findings were unremarkable. Relevant laboratory investigations were requested. Empirical antibiotic therapy was commenced using intravenous (IV) ciprofloxacin and metronidazole. There was no bacterial growth on blood cultures. Oral antimalarial treatment was administered following a positive blood film for Plasmodium falciparum. Further supportive care included antipyretics and IV fluid.

Two days later, he was reviewed by the consultant physician who documented a positive history of haematuria, subconjuctival/buccal mucosal haemorrhage and clinical deterioration. A diagnosis of suspected VHF was made. He was isolated and the case was notified. Due to lack of local VHF diagnostic and isolation facilities and considering that ribavirin was not readily available in Enugu State, he was moved to the ILFRC, Irrua, on 12 October 2014 where diagnosis of LF was confirmed. He was admitted into the LF isolation facility and commenced on ribavirin therapy and supportive care. The following day, he had a session of haemodialysis but eventually died of multi-organ failure.

Contact tracing

The ICST identified 121 primary contacts. Their demographic and epidemiological characteristics are shown in Table 1. Their mean age was 32.8 ± 6.6 years, and the majority (65.3%) were women. Seventy-eight (64.5%) contacts were hospital staff/interns.

Risk categorisation

The details of risk categorisation are summarised in Table 2. Only 20 contacts (16.5%) were classified as high-risk.

Clinical outcomes of monitoring of primary contacts

Over the 21-day period of monitoring, three persons developed fever all of whom belonged to the high-risk group. The first case developed fever on the fourth day post-exposure and before initiation of ribavirin. Lassa virus RT-PCR and malaria parasite film were performed both of which were negative. He was empirically administered oral ribavirin therapeutic regimen, antimalarial and antibiotics and he stabilised. The second case developed fever on the fifth day post-exposure and before initiation of ribavirin. He declined requested laboratory investigations but was offered antimalarial and empirical ribavirin therapy following which he became afebrile. The last case developed fever on the fifteenth day post-exposure, which was after completion of ribavirin PEP. His laboratory findings included falciparum malaria parasitaemia and neutrophilia. Lassa virus RT-PCR could not be performed due to logistic reasons. He became afebrile after antimalarial/antibiotic treatment.

Laboratory abnormalities in primary contacts

Only 37 contacts (16 with high-risk and 21 with low-risk exposure) underwent laboratory investigations for surrogate markers. The abnormalities observed are summarised in Table 3. The commonest abnormalities were prolonged international normalised ratio (INR) seen in 17 (45.9%), raised total bilirubin seen in 16 (43.2%) and lymphocytosis observed in 12 (32.4%) persons. When contacts with high-risk and low-risk exposures were compared, statistically significant differences were observed only in the proportion of persons with prolonged INR (68.8 vs. 28.6%, 95% CI 1.10–30.03, P = 0.02), raised total bilirubin (75.0 vs. 19.0%, 95% CI 2.15–88.17, P < 0.01) and elevated aspartate transaminase (25.0 vs. 0.0%, P = 0.03).

Ribavirin post-exposure prophylaxis

Eighteen of the 20 persons with high-risk exposure were prescribed ribavirin. One person did not consent to ribavirin PEP, while another was ineligible due to pregnancy. Of the 18 persons prescribed ribavirin, 13 had 100% adherence and three had 90% adherence, while the remaining two people did not take the drug.

The side effects reported by those who received ribavirin are presented in Table 4. There were 25 events reported in 12 persons. Fatigue and dizziness were the commonest side effects and were reported in 7 (43.8%) and 5 (31.3%) persons, respectively. None of the contacts considered the side effects severe enough to discontinue the drug.

Psychosocial challenges of primary contacts

The commonest psychosocial challenge was stigmatisation of primary contacts by their colleagues. Two persons with high-risk exposure reported depression not related to ribavirin use. Abstinence from sex as an infection control measure at least within the 21-day incubation period also constituted a challenge.

Achievements

The case was recognised as suspected VHF within 48 h of presentation to our hospital which led to institution of prompt response activities. The success of the response team largely hinged on three points. First, following the importation of Ebola into Nigeria, some modalities had been put in place as part of Ebola preparedness. We also benefitted from our previous experience in contact tracing and management of primary contacts of a case of LF that spilled over to our hospital from a neighbouring state earlier in the year. Moreover, the hospital management was supportive and easily accessible.

The ad hoc establishment of a section of the laboratory with improved containment for routine laboratory investigations for suspected VHF patients or their primary contacts is another achievement worth mentioning. This made it possible to run the investigations at minimal risk to laboratory staff.

The hospital also converted a previously underutilised unit to a small infectious diseases isolation and treatment facility. This facilitated the management of primary contacts and minimised unnecessary panic which would have resulted from care of primary contacts in the general clinic.

Gaps and challenges

A number of gaps/challenges were identified. First, the patient was initially admitted with a working diagnosis of sepsis. As has been observed previously 20, 22, it is common for diagnosis of VHF to be missed especially in the early stages of the disease when the clinical features are indistinguishable from other febrile illness. Clinicians in sub-Saharan Africa may therefore be justified when they exercise prudence to rule out commoner and more treatable infections in such patients. Moreover, the apparent lack of VHF exposure history and the attempt by the patient to conceal haematuria made earlier suspicion of VHF difficult. Although the presence of bleeding may help to discriminate between VHFs and some other causes of fever, overdependence on bleeding tendency may be misleading as it occurs in only about 30% of cases of LF and in 40–50% of Ebola patients and it may not be externally visible 2, 23.

The second gap we identified was the failure of a reasonable number of health workers to use appropriate personal protective equipment (PPE) during close contact with the index case. This is worrisome because elaborate health education and sensitisation on the prevention and management of VHF had been conducted by the hospital and government agencies following the importation of Ebola into Nigeria.

Thirdly, the absence of a purpose-built VHF isolation facility in the hospital was a major challenge. In the absence of this, makeshift isolation was provided following the diagnosis of suspected VHF, but this did not provide the level of containment required for optimal patient care at minimal risk to health workers.

As was evident in our experience, lack of facilities for laboratory diagnosis of Lassa virus infection within short distances of endemic regions has been identified as a major challenge in VHF management in Nigeria 22. This prevented more swift decisions on the index case and beclouded optimal evaluation of symptomatic primary contacts. Only one of three symptomatic primary contacts had Lassa virus RT-PCR test due to logistic reasons. Although the RT-PCR was negative for Lassa virus, a repeat investigation few days later would have been carried out if the facilities were readily available. A positive Lassa virus RT-PCR following an initial negative result has been previously documented in a primary contact of LF in south-east Nigeria 22. Although Lassa virus infection could be diagnosed by enzyme-linked immunosorbent serologic assays (ELISA) which detect IgM and IgG antibodies as well as Lassa antigen, the fact that antibodies may be undetectable in the early stage of the disease limits their utility. Lassa virus-specific RT-PCR is therefore preferred for diagnosis in the early stage of the disease. Lassa virus serologic assay was not performed for any of the primary contacts due to logistic reasons. However, serological assessment 2–3 weeks after exposure to the index case would have elucidated the risk of infection following various levels of exposure to Lassa virus and further helped to assess the role of ribavirin PEP in those with serological evidence of infection. The lack of serological assessment is a limitation in this study.

Availability of ribavirin on demand has frequently been a major challenge in LF management in many Nigerian hospitals. As the ribavirin stock of the Enugu State Directorate of Public Health had expired, ribavirin was only initiated after the index case arrived at the National Reference Centre. This further explains why we waited for 48–72 h to procure ribavirin administered to eligible primary contacts.

Lessons learnt

A number of lessons were learnt most of which could be useful in the future. Health workers need to have a high index of suspicion for VHF especially in endemic communities. Periodic booster sensitisations and drills may be necessary to optimise health workers’ approach to VHF care. While VHFs may be clinically indistinguishable, in African communities endemic for LF and where endemic transmission of Ebola is strange, empirical ribavirin therapy for LF following strong suspicion of VHF may be justifiable. There is need for improved communication between clinicians and laboratory staff. There is a pressing need to increase the availability of ribavirin on demand in localities where LF is endemic and in neighbouring referral hospitals. Government and non-governmental partners including WHO should make this possible. Establishment of LF diagnostic and treatment centres within short distances of endemic communities is a challenge that we must overcome. Stigmatisation of VHF contacts needs to be aggressively confronted.

In conclusion, we described a case of LF that occurred in Nigeria during a devastating outbreak of Ebola in West Africa, thereby putting the healthcare team at a crossroads. In the course of management of the patient, several individuals were exposed. Well-coordinated infection control/surveillance strategies were instituted. Ribavirin was offered to high-risk contacts as post-exposure prophylaxis. No secondary case of LF occurred. While we call on governments and non-governmental organisations to step up the fight against the ongoing scourge of Ebola in West Africa, our experience underscores the need to comprehensively strengthen the weak health systems of sub-Saharan Africa; otherwise, we might win the Ebola battle but lose the one against less virulent infections for which effective treatment exists.