Verbal consent was sought from the close relatives of the patients. The officers were also spoken to personally by the medical team at the 37 Military Hospital.

The index case was a male soldier with severe symptoms of malaria who died on May 11, after admission to a Level II hospital for about a week. Two male colleagues, aged 27 years (Patient 1) and 33 years (Patient 2), who were living in the same camp, subsequently developed severe fever and myalgia on May 15 (Patient 1) and 21 (Patient 2). Their medical conditions deteriorated, necessitating their medical evacuation to the 37 Military Hospital, a Level IV facility, in Accra, Ghana, on May 26, 2013 after five days of illness for Patient 1 and 11 days for Patient 2. Patient 1 died two days after admission. At the military hospital in Accra, the medical staff ruled out malaria through microscopic examination of thick and thin blood smears and focused on the clinical manifestations, which were classified as VHF. A presumptive clinical diagnosis of Lassa fever was made on the strength of previous outbreaks of Lassa fever in Liberia.¹⁰ To investigate these clinical suspicions, blood samples from the patients and five asymptomatic close contacts were analysed.

Five millilitres of blood were collected from each of the two patients (Patient 1 sample number: 60FSD_28052013; Patient 2 sample number: 59_PN_28052013) and five asymptomatic contacts (medical staff who had initially attended the two patients without barrier nursing). The samples were transported to the Virology Department of the Noguchi Memorial Institute for Medical Research at the University of Ghana in Legon, Ghana. The blood samples were processed into serum by low-speed centrifugation at 4 °C and the resultant 3 mL aliquots of sera were cryopreserved at -80 °C.

Viral RNA was extracted from 140 µL of the blood samples using the QIAamp viral RNA mini kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. A diagnostic assay developed by Panning et al.,¹¹ for filovirus species and carried out with a OneStep RT-PCR reaction kit (Qiagen, Hilden, Germany) was used in a 25 µL total reaction volume, including 3 µL of the extracted RNA. The real-time diagnostic assay used five optimised large (L)-gene primers and three probes, as well as an internal control with a separate detection probe. Reactions were supplemented with 40 ng/mL bovine serum albumin and 400 mmol/L each dNTP. The primers and probes used were designed and published by Panning et al.¹¹ (Table 1). All probes were synthesized by Tib-Molbiol (Berlin, Germany). Amplification in a 96-well ABI 7300 Real Time PCR System (Life Technology Holdings, Singapore) involved the following steps: 50 °C for 30 minutes; 95 °C for 15 minutes; and 45 cycles of denaturation at 95 °C for 15 seconds, annealing at 52 °C for 25 seconds and extension at 72 °C for 20 seconds. Fluorescence was measured at the end of each 52 °C annealing step.

An endpoint reverse transcription (RT)-PCR protocol for the detection and identification of flaviviruses, developed by Pierre, Drouet and Deubel in 1994¹² with a set of universal oligonucleotide primers, was used for laboratory investigation of flaviviruses present in the serum samples. These primers correspond to the 3’ non-coding region of the NS5 gene, which is highly conserved amongst the mosquito-borne flaviviruses. The OneStep RT-PCR kit (Qiagen, Hilden, Germany) was used with the following cycling conditions: reverse transcription at 50 °C for 30 minutes; initial activation step at 95 °C for 15 minutes; 45 cycles in a 3-step cycling of 95 °C for 30 seconds, 52 °C for 30 seconds and 72 °C for 30 seconds, with a cooling step of 30 seconds at 30 °C.

A gel-based conventional RT-PCR assay was performed using ABI 2720 Thermal Cycler (Life Technology Holdings, Singapore) for the detection of Lassa virus with primers specific for regions of the S RNA segment. The RT-PCR (45 cycles) contained OneStep RT-PCR kit reagents (Qiagen, Hilden, Germany) with the sense primer 36E2 and the antisense primer LVS-339-rev, as described by Ölschläger et al.¹³ (Table 1). Cycling conditions for the RT-PCR involved the following steps: 50 °C for 30 minutes; 95 °C for 15 minutes; and 45 cycles of 95 °C for 15 seconds, 52 °C for 30 seconds and 72 °C for 30 seconds. The amplification products (expected size: 320 bp) were electrophoresed on a 2% agarose gel (peqlab Biotechnologie, Erlangen, Germany), stained with ethidium bromide and viewed under a Kodak Gel Logic 100 Imaging System (Cole-Parmer Int., Chicago, Illinois, United States).

An RT-PCR assay specific to the L-gene segment of the Lassa arenavirus on the RNA extracted from the processed blood samples. The 45 cycle RT-PCR used a AgPath-ID One-Step RT-PCR kit (Ambion, Life Technologies, Thermo Fisher Scientific, New York, New York, United States) with the primers as described by Vieth et al.¹⁴ (Table 1). The reaction solution consisted of: 2.5 μL nuclease-free H₂O, 12 μL 2X RT-PCR buffer, 2 μL of the upstream and downstream primers (10 μM each), 1.5 μL 2X RT-PCR buffer enzyme mix and 5 µL DNA. Nucleic acid amplification started with a 30 minute RT step at 50 °C, then an initial denaturation step of 2 minutes at 95 °C and a 45-cycle amplification of 20 seconds at 98 °C, 30 seconds at 55 °C and 60 seconds at 72 °C. A gel was used to separate the products of the nucleic acid amplification. Images of the DNA bands were captured using a Gel Logic 100 imaging system (Eastman Kodak Company, Rochester, New York, United States). The DNA bands of nearly 400 bp in size were analysed and purified for sequencing.

A real-time RT-PCR assay was performed using the ABI 7300/7500 Real Time PCR System (Life Technology Holdings, Singapore) for the detection of Lassa virus with primers specific for regions of the small (S) RNA gene segment. The reagent used in the preparation of the master mix for the 45-cycle RT-PCR was the Power SYBR® Green RNA-to-CT™ 1-Step Kit (Life Technologies, Carlsbad, California, United States) with the sense primer 36E2 and the antisense primer LVS-339-rev, as described in detail by Ölschläger et al.¹³ (Table 1). The amplification protocol was as follows: 48 °C for 30 minutes; 95 °C for 10 minutes; and 45 cycles of 95 °C for 15 seconds, 60 °C for 1 minute. At the end of each cycle, fluorescence was measured at 60 °C.

The PCR products generated in the conventional RT-PCRs from the L-gene segment fragments were purified in accordance with the manufacturer’s instructions using a commercial kit (Big Dye XTerminator Purification Kit, Applied Biosystems, Carlsbad, California, United States). The purified products were sequenced on both strands using the L-gene PCR primers and an ABI Prism 3130 genetic analyser (Hitachi High Technology, Singapore). The sequences were assembled with Lasergene software (DNASTAR, Thermo Fisher Scientific, New York, New York, United States) and automated base calling was proofread by visual inspection of the electropherograms. Other published Lassa virus sequences were obtained from the National Center for Biotechnology Information’s^15,16 GenBank (Table 2) to align with the generated sequences for phylogenetic analysis. With the use of the National Library of Medicine’s Basic Local Alignment Search Tool¹⁷ (BLAST) program, we compared the patients’ Lassa virus sequences with other reference Lassa virus sequences to locate regions that were similar. All sequences obtained were copied into BioEdit software (North Carolina State University, Raleigh, North Carolina, United States) and put together into an ideal sequence alignment file with the use of a multiple sequence alignment tool, ClustalW¹⁸ (version 2; EMBL-EBI, Cambridge, United Kingdom). We then used Molecular Evolutionary Genetics Analysis (MEGA) software (version 5.05) in a Kimura two-parameter model¹⁹ to infer a phylogenetic tree from the aligned nucleotide sequences following a neighbour-joining method. To establish reliability and to infer the strength of similarity between the patient’s sequence and the referenced sequences from the phylogenetic tree, a bootstrap analysis of 1000 replicates was used.

The gel-based RT-PCR flavivirus tests for Yellow fever and Dengue fever (types 1–4), as well as the real-time RT-PCR filovirus test for Marburg and Ebola, were negative (Table 3). However, the RT-PCR amplification of the S-gene segment of arenaviruses detected a 320 bp DNA band in the well with the patients’ clinical specimens. The length and position of these bands were on a par with the positive Lassa virus control (an inactivated culture supernatant of cells infected with Lassa virus strain CSF); no band was observed in the negative control (PCR-grade water) well. The real-time RT-PCR amplification of the S-gene segment of the arenavirus test produced a clear peak with a sigmoid-shaped curve for the patient’s samples, whereas no peak was observed for the negative control, which in the figure is covered by the threshold line, or baseline (Figure 1). This two-result signal indicated the presence of Lassa virus in the patients’ sera. The samples from the five asymptomatic close contacts were tested in the same assay run and no indication of the presence of Lassa virus was observed (data not shown).

The gel-based RT-PCR amplification assay for the large L-gene segment of the arenavirus showed ~400 bp products that matched the expected size (Figure 2). The comparison of the nucleotide sequences from the amplified products with the GenBank database showed that the patients’ sequences had high similarities to known Lassa virus strains. Genetically, the sequences were close to strains that had originated in Guinea and been reported in Liberia. The highest similarity (85% maximum identity in nucleotides) was to Lassa virus strain Macenta²⁰ and Lassa virus strain las803792.¹⁴

The phylogenetic analysis with GenBank data revealed that the nucleotide sequences from the patients had close phylogenetic relationships with the reported Lassa virus strains Macenta²⁰ (GenBank accession number: AY628200) and las803792¹⁴ (GenBank accession number: AY693638) (Figure 3). The sequence data from the patient who died (Patient 1, sample no. 60FSD_28052013) was sent to GenBank and assigned accession number KF425246.

Our report was limited by our inability to conduct a battery of tests, including serological assays, for either an ideal suspected case(s) of VHF or for the five asymptomatic close contacts of the evacuated patients. That notwithstanding, this report underscores the importance of preventive measures for all visitors and workers including peacekeeping forces to endemic regions in West Africa. This is because there is currently no effective Lassa fever vaccine is available.

It is recommended that medical support plans for peacekeeping operations should be built purposefully and in consideration of existing endemicity and history of endemicity in the host nations. Such support plans should be duly informed by frequently-updated research on endemic agents and other health concerns in the visiting country. Sensitisation of officers and other travellers to infectious agents in the host country and the importance of speedy reporting to health facilities when unwell should be given sufficient emphasis. This report of laboratory investigations of imported cases of Lassa fever and other documented medical fatalities on past peacekeeping operations supports the need for essential medical organisational changes in future operations. This would involve a good balance of proximity to medical care and transportation time for medical emergencies.

In conclusion, it is envisaged that the importation of VHFs into non-endemic countries will increase in likelihood as a result of increased travel and international commitments in conflict and disaster situations to VHF-endemic countries in West Africa. Thus, healthcare providers should: (1) have a high index (low threshold) of suspicion for VHF amongst travellers returning from endemic areas; (2) promptly implement appropriate infection prevention and control measures; and (3) rapidly report suspected cases to avert undue nosocomial transmission.