James E. Strong. NML, PHAC, 1015 Arlington St, Winnipeg, MB, Canada, R3E 3R2. Tel.: 204 789 7032; Fax: 204 789 2140; E-mail: firstname.lastname@example.org and Heinz Feldmann. Laboratory of Virology, NIH, 903 South 4th St, Hamilton, MT 59840, USA. Tel.: 406 375 7410; Fax: 406 375 7416; E-mail: feldmannH@niaid.nih.gov
The mobile laboratory provides a safe, rapid and flexible platform to provide effective diagnosis of Ebola virus as well as additional differential diagnostic agents in remote settings of equatorial Africa. During the 2007 Democratic Republic of Congo outbreak of Ebola-Zaire, the mobile laboratory was set up in two different locations by two separate teams within a day of equipment arriving in each location. The first location was in Mweka where our laboratory took over the diagnostic laboratory space of the local hospital, whereas the second location, approximately 50 km south near Kampungu at the epicentre of the outbreak, required local labour to fabricate a tent structure as a suitable pre-existing structure was not available. In both settings, the laboratory was able to quickly set up, providing accurate and efficient molecular diagnostics (within 3 h of receiving samples) for 67 individuals, including four cases of Ebola, seven cases of Shigella and 13 cases of malaria. This rapid turn-around time provides an important role in the support of patient management and epidemiological surveillance.
• During the 2007 outbreak of Ebola-Zaire in the Democratic Republic of Congo (DRC), a mobile laboratory was set up in two separate locations, the first in an existing building (a wing of the local hospital in Mweka) and the second in a newly fabricated tent structure near the epicentre of Kampungu, both being capable of running within <24 h of arrival.
• The mobile laboratory provided safe, accurate, rapid diagnostic quantitative real-time PCR (Q-RT-PCR) testing of Ebola virus as well as differential diagnostics for selected enteric pathogens and malaria which were comorbid.
• As these outbreaks typically occur in remote settings of equatorial Africa, not well-serviced by roads or air, with inconsistent or non-existent infrastructure, the mobile field laboratory offers rapid and reliable diagnostic services assisting in the management of the outbreak and depending on only minimal local and non-governmental organization support.
The filoviruses, Ebola and Marburg, have caused sporadic outbreaks of viral haemorrhagic fevers in geographically isolated areas of equatorial Africa over the past 30 years (Groseth et al., 2007). Fatality rates for filovirus infections in humans range from 23% to 90% and despite the fact that there are no effective treatments or prophylaxes available, effective case management requires discriminating cases of filovirus haemorrhagic fevers (FHF) from other, less severe and potentially treatable conditions that may coincide (Feldmann and Geisbert, 2011). As there are no pathognomonic features of FHF clearly identifying cases from non-cases, in the absence of laboratory diagnosis, medical management, isolation, contact tracing and surveillance must be resourced to all suspected cases. Furthermore, because these outbreaks typically occur in settings that are remote, having few manpower and infrastructural supports, economical utilization of these resources becomes paramount in the early quelling of the outbreak. Filovirus outbreaks also present significant health and social problems for the affected communities that are already underserviced with medical resources. Many of these problems can be more effectively managed through timely and accurate diagnosis of suspect cases. The ability to provide efficient diagnosis via a mobile field-based laboratory, with rapid turn-around times enables trained surveillance, isolation and treatment teams to concentrate their efforts in the areas of most concern, thereby allowing existing health care teams to manage the patients that are harbouring less lethal pathogens (Roddy et al., 2007; Grolla et al., 2011).
During the September 2007 outbreak of Ebola Haemorrhagic Fever in the Democratic Republic of the Congo (DRC) (Fig. 1), the National Microbiology Laboratory (NML) of the Public Health Agency of Canada (PHAC) mobile field laboratory was deployed to provide diagnostic services for Ebola as well as differential molecular diagnostics for selected enteric pathogens and malaria. PHAC-NML offered assistance to the World Health Organization (WHO) as a partner of the ‘Global Outbreak Alert and Response Network’ (GOARN). Under GOARN, the mobile laboratory was first deployed to Mweka (September 20–October 10) and then to Kampungu (the epicentre of the outbreak from October 10 to 24), to assist clinical management and epidemiological surveillance. The outbreak was centred in a remote area of the Kasai Occidental province near Kampungu, a village approximately 150 km (93 miles) north-west of the regional capital Kananga in the DRC (Fig. 1). This area is not well-serviced by roads or air and presented unique sets of logistical/technical problems that the laboratory had not previously encountered to this extent. Here, we present a summary of the laboratory’s efforts and approaches to manage the response.
People in this area initially presented in April, 2007 with Ebola-like symptoms including fever, myalgia, vomiting, diarrhoea and some with haemorrhage. The presence of the disease was not confirmed until September, 2007 by the Centers for Disease Control and Prevention (CDC) in Atlanta whom also brought their mobile laboratory to assist in management near Kampungu (World Health Organization, 2007). Experts from WHO and Medecins Sans Frontieres (MSF) were mobilized and brought in medical equipment, sanitation products and personnel to help deal with the outbreak. MSF and WHO undertook surveillance activities looking for suspect cases in villages within a 30 km radius of the epicentre of Kampungu. Clinical samples were collected by personnel wearing personal protective equipment (PPE), including a surgical mask, cap, shield or goggles, gown, apron, gloves (two pairs) and boots. Whole blood and serum samples were collected from most patients using EDTA and serum Vacutainers® (BD, Franklin Lakes, NJ, USA), respectively. Swab samples (nasal and oral) were collected from one deceased individual using cotton-tipped applicators (AMG Medical, VWR, Mississauga, ON, Canada) and stored in 700 μl of Dulbecco’s modified essential medium (DMEM) supplemented with 5% bovine serum albumin (Invitrogen, Burlington, ON, Canada). Eighteen faecal samples were provided as swab samples or directly in sample cups. For transport to the field laboratory, tubes or cups were sealed in plastic bags, surface disinfected with a 1% hypochlorite solution, sealed into a second bag or container and again surface disinfected. Collection of human specimens occurred on a WHO outbreak response protocol.
Field molecular diagnostics was provided by the PHAC-NML mobile laboratory that utilized a negative pressure flexible film isolator (see Fig. 2) to safely handle and inactivate infectious agents and Q-RT-PCR assays to detect the nucleic acid of targeted pathogens. Samples were manipulated in the isolator using additional personal protective equipment (PPE- including N95 mask, face shield, a double layer of nitrile gloves, TyvekTM (Dupont, Wilmington, DE, USA) gown). For testing, an aliquot (140 μl of blood, serum or swab storage solution) was removed from each sample and inactivated by adding 560 μl of the lysis buffer AVL (Qiagen, Mississauga, ON, Canada). The sample tubes were submerged in 1% hypochlorite solution in the airlock of the isolator that was thoroughly sprayed with the same solution and allowed to decontaminate for at least 10 min prior to removal. All further work was performed with PPE as outlined above on the bench. Nucleic acid isolation was completed using the QIAamp Viral RNA mini kit (Qiagen) as according to the manufacturer’s instructions except for the addition of a second AW1 wash to ensure no PCR inhibiting substances were copurified along with the nucleic acids. All waste material was treated with 1% hypochlorite solution and incinerated on the same day. Aliquots of each sample were prepared for transportation to the Institut National de Recherche Bio-Médicale (INRB) in Kinshasa for storage. Transportation was carried out in compliance with International Air Transport Association (IATA) regulations after prior approval by the appropriate national authorities of the sending or receiving countries. Reagents and the laboratory team were replaced after 3 weeks which coincided with the movement of the laboratory.
Two Ebola Q-RT-PCR assays were used that targeted regions of theVP40 gene (ZEBOVVP40F – 5′-ccaaaacttcgccccattc, ZEBOVVP40R – 5′-cggcactgttccccttctt, ZEBOVVP40P – 6FAM-tttacccaacaaragtg-MGBNFQ) and the polymerase (L) gene (ZEBOVLF – 5′-gcgccgaagacaatgca, ZEBOVLR – 5′-ccacaggcacttgtaacttttgc, ZEBOVLP – 5′-6FAM-tggccgccagcct-MGBNFQ) of Zaire ebolavirus. Additionally, testing was carried out for Shigella dysenteria, Salmonella typhi, Vibrio cholera and Plasmodium sp. (Wang et al., 2007; Lyon, 2001; Lee et al., 2002). Detection was carried out using the Lightcycler 480 RNA Master hydrolysis probes kit (Roche, Laval, PQ, Canada). Briefly, 5 μl of RNA was added to 20 μl of master mix containing 1× reaction buffer, 3.25 mm Mg[OAc]2, 0.5 μm forward and reverse primers, 0.25 μm probe and 1× enhancer solution. Q-RT-PCR assays were run on Smartcycler thermocyclers (Cepheid, Sunnyvale, CA, USA) using the cycling profile 61°C for 300 s, 95°C for 30 s followed by 40 cycles of 95°C for 15 s and 60°C for 40 s. A single acquisition of fluorescence was taken at the end of each cycle (Fig. 3).
The two PHAC-NML mobile laboratory teams were deployed to the DRC in succession; the first set up in two adjacent rooms of an existing local hospital in Mweka (Fig. 2a) and then a second in a newly constructed tent on the MSF compound in Kampungu (Fig. 2b) both in Kasai Occidental province. In both settings, the laboratory was able to quickly set up (within a day of equipment arrival), providing accurate and efficient molecular diagnostics (within 3 h of receiving samples) for 67 individuals, including four cases of Ebola and seven cases of Shigella (all tested at the Mweka site) and 13 cases of Malaria (eight tested at the Mweka and five at the Kampungu site) (Table 1). While we were set up to perform serologic testing in both settings, this did not represent a significant portion of the testing performed. In fact, no serologic testing was performed at the Kampungu site at all and serology was not part of the formal diagnostics performed in this outbreak mission. Both teams consisted of three individuals, although the workload can be effectively managed with only two personnel; (i) a research technician, whom is proficient with handling, isolation, inactivation, extraction and running of RT-PCRs, and (ii) a scientist whom is adept at all of the above as well as the interpretation of diagnostic tests.
Table 1. Field laboratory Q-RT-PCR testing results
Sample (no. of patients)
ND, not done.
The number of patients positive in testing for each pathogen and each sample source is shown.
aSwab samples (nasal and oral) were collected from one deceased individual.
Oral/nasal swab (1)a
For the second deployment, a suitable existing structure was not available in the Kampungu area. The local ambulatory care clinic was offered by local medical officials; however, in discussion with MSF and WHO partners, this was rejected as we did not wish to disrupt existing, important health programs (including vaccination, health supervision and prenatal care) during our tenure. Another structure was offered by local officials and that being the one vacated by the CDC-Atlanta team. The departure of one laboratory was the result of a joint decision among all participating parties as it seemed obvious that one laboratory was sufficient to support the remaining outbreak investigation. However, rising security concerns precluded the use of this space given that it was remote from both the MSF and WHO sites. As such, a tent structure was built using local builders and materials (Fig. 2b). The choice to put this on the MSF compound was done for logistical reasons, given its proximity to the isolation unit and for aforementioned security reasons. The interior laboratory area was divided up with the use of tarps for separation of work areas (Fig. 2b).
The mobile laboratory is relatively self-sufficient requiring only minimal local and/or non-governmental partners to supply a building or shelter to locate the laboratory, in-country transport, security and logistic support, as well as food and a supply of water. Remaining supplies and equipment are provided in our mobile laboratory set up including a Class III flexible film glove box, two real-time thermocyclers (SmartCycler®, Cepheid, Sunnyvale, CA, USA), molecular assays and reagents for the outbreak virus and selected differential agents (sufficient for 1000 tests), as well as all equipment and supplies necessary for all sample handling and testing (e.g. centrifuge, pipetters, PPE, consumables, disposable labware) (see supporting information). We were also able to place an incinerator just outside the laboratory area to safely dispose of laboratory waste (not shown). Both laboratory set-ups were completely powered by a portable 1000 watt generator that we supply.
The Q-RT-PCRs we utilized during this outbreak were probe-based with fluorescent tags that could be run as multiplex reactions, simultaneously testing one sample for multiple pathogens (Fig. 3). In fact during this outbreak, we were able to simultaneously detect cases of Malaria and Ebola in the same patient. It is unclear at present whether these co-infections were clinically related or rather represent the high background prevalence of malaria in the population affected. Certainly, co-infection with Ebola virus might impact the level of parasitaemia in a patient infected with malaria given this virus’ known effects on adaptive immunity, lymphopenia and effects on interferon activities (Plebanski and Hill, 2000; Feldmann and Geisbert, 2011). Suitably designed probe-based assays can provide increased sensitivity and specificity when compared to other assays (Towner et al., 2004) and can be easy to interpret, however, with new or uncharacterized strains, SYBR green assays can be more reliable as only two primers to well conserved regions of the viral genome need to be present. It remains unclear how many of a total of 264 suspected cases were due to Ebola because we nor CDC-Atlanta tested most of these. Of 110 samples taken from suspected Ebola victims, a total of 26 tested positive for the disease during this combined surveillance. We suspect that shigellosis was an important factor during this outbreak as seven of 18 stool samples we received were positive for this pathogen.
As in previous outbreaks, the inclusion of laboratory support to assist in the allocation of manpower and resources substantially helps these efforts. With rapid turnaround time, capable of providing diagnoses for a specific pathogen within 3 h of having obtained the specimen, a mobile field laboratory has the biggest impact in the following areas: (i) Appropriate isolation can be initiated for positive cases, minimizing unnecessary use of limited isolation resources, (ii) Contact tracing can be initiated for confirmed cases that are often very labour- and resource-intensive tasks and as such limited resources are not wasted on non-cases, (iii) Identification of positive cases among community deaths, thereby allowing traditional burial rites where cases test negative and ‘safe’ burials in positive cases with contact tracing initiated and (iv) Aid in re-acceptance of negative or convalescent cases into community with negative laboratory testing results.
Offering other tests including differential diagnostics significantly increases the value of the on-site laboratory. This was much harder to achieve in the field and requires variable clinical specimens (in particular blood or stool), more manpower and more extensive and continuous supplies. At a minimum, malaria diagnostics (e.g. commercially available rapid dipstick tests and/or PCR-based testing) should be included for communicable FHF outbreaks in malaria-endemic areas. Diagnostics for severe gastrointestinal infections are also often useful as was the case here. Differential diagnostic testing allows for treatment to be implemented for these more treatable diseases. Additional laboratory testing could also be implemented in the field including blood chemistry, gases and haematology analysis, which would substantially aid physicians in supportive care management of the FHF as well as other pathogen-infected patients.
Typically, teams are deployed into the field for 3–4 weeks total duration each. As filovirus outbreaks require active surveillance for up to 21–42 days beyond the last positive case, the deployment of at least two functional teams is necessary. In this outbreak, only two teams were deployed; however, this is the first time in which the mobile laboratory was packed up by one team, moved to a new location and set up by the second team. This was also the first time in which there was no overlapping of teams during the transition. Although there were complications that occurred, including the temporary loss of equipment, supplies and personal effects by the international airline for the first team, laboratory set-up was accomplished in both settings within 24 h of eventual equipment arrival.
The usefulness of on-site laboratory support has been questioned in the past, including some concern about the sensitivity and specificity of the assays during early disease stages and for survivors in the early convalescent stage. A previous outbreak of Marburg virus in Angola (CDC, 2005) demonstrated that this laboratory can be effective and efficient in several aspects, providing substantial benefits to outbreak response management (Towner et al., 2006; Grolla et al., 2011). The results from this outbreak demonstrate that rapid turn-around times can clearly aid in surveillance and case load management. Concerns have also been raised about costs, security, health risks and other potential dangers associated with outbreak missions. Even though risks are always present, they have been mitigated by careful planning and collaboration with non-governmental organizations such as WHO and MSF and have not manifested during these or other outbreak missions.
We would like to thank the individuals and agencies that aided us in our response, especially Medecins Sans Frontieres (MSF), the World Health Organization, the National Microbiology Laboratory of the Public Health Agency of Canada, the DRC Ministry of Health, the Special Pathogens Branch of the Centers for Disease Control and Prevention (CDC), Atlanta and the Canadian Embassy (Kinshasa). Our sincere thanks go to the communities affected by the Ebola outbreak. We highly appreciated the warm welcome and the local support. Further, we would like to thank the regional and community authorities of the Democratic Republic of the Congo for their support.
Conflicts of interest
The authors have no potential conflicts to declare.