Last month, Tom and I (Matt) visited Becton Dickinson's instrument test laboratory at their impressive Erembodegem premises, just outside of Brussels. Again, Janet Midega from Wellcome joined us for some of the trip. We were initially hosted by Roeland Bernaerts and Roel Van Cauter, who gave us detailed technical assistance in hooking up to the BACTEC FX blood culture system and the Phoenix M50 ID/AST analyser, our two primary targets. We tested against both instruments via EpiCenter, BD's entry level middleware, and direct to each instrument, as we are likely to encounter both scenarios in target deployments. By the end of the first day, we had achieved successful bi-directional communication in all scenarios, due in no small part to Roel and Roeland's attentive support (thanks guys!).
As can be seen from the above images, the lab also featured a Mycobacteria Growth Indicator Tube (MGIT) system and a MAX molecular diagnostic analyser. Given all BD instruments support the CLSI LIS01 and LIS02 protocols for LIS/LIMS interaction, we decided to have a go at interfacing with these systems also, as a bonus! On the SEDRI-LIMS side, we have implemented these protocols in such a way that new systems can potentially be supported simply by configuring new mappings to and from internal fields (which themselves can be configured as part of new direct or culture tests). Once the appropriate mappings were in place, we gave it a whirl. Success! While we did not have the expertise (or likely materials) on hand to run the susceptibility phase of the MGIT operation, we were able to simulate detection of Mycobacterium and correctly receive the corresponding results at the LIMS. Similarly with the MAX, we were able to receive a range of simulated results at the LIMS.
The Brucker BioTyper (a MALDI-ToF) was unfortunately not available for testing against. Moreover, although this also supports EpiCenter and the LIS01/LIS02 protocols, precise details of the message content were hard to pin down. However, Roeland eventually managed to extract the relevant documentation for us from a colleague at Brucker and, while we've yet to test against a real system, we do now have the mappings in place.
For EpiCenter supported instruments, it became clear that, as far as Roel and Roeland were concerned, the ability of a LIMS to correctly receive and process EpiCenter messages sourced from its extensive database of recorded instrument responses, in lieu of the corresponding instrument being physically connected, was tantamount to supporting that instrument. Hence while we were able to test against real versions of at least the BACTEC FX, Phoenix M50, MGIT and MAX, it would clearly be of huge benefit to us to have the latest database available for regression testing back at the office (and for high confidence conformity testing against new instruments or instrument versions prior to a physical connection). To this end, Roeland very kindly gave us a copy of the full database to take home with us! It includes Brucker BioTyper responses which we have since demonstrated our support for.
On Wednesday, we met up with Rudy Bastin, Senior Manager for Business Development and Global Health. We've known Rudy for a couple of years now, since we first started working with BD in fact, but this was our first face to face encounter. It was great to finally meet Rudy in person and catch up properly, including over dinner Wednesday and Thursday evening (thanks Rudy!). Someone else we met on Wednesday was Lut Van Den Broeck, an application specialist, who took us through the typical end-to-end workflows for the various instruments of interests. This was invaluable information and, I'm happy to say, we later received a detailed document on the subject which we were not previously aware of.
On Friday morning, shortly before our departure, we had a wash-up meeting with Rudy, Roel, Roeland and Mohamed (El Makrani). We reflected on a hugely successful week and discussed next steps, one of which was to set up an end-to-end demo of all SEDRI-LIMS / BD instrument workflows that we had managed to prove during the week, this time with the superset of global BD staff that we have been working with over the last couple of years, including Parul Goyal, Nuphar Rozen-Adler, Diane Flayhart, Roger Nicolson and others. With the AST aspects of the MGIT operation and a physical test against the Brucker still outstanding, a future trip to another BD site such as Heidelberg was proposed. There was also enthusiasm from both sides for going beyond BD's suite of microbiology specific instruments, given that SEDRI-LIMS now covers other laboratory disciplines, such as haematology. For example, BD's flow cytometer was proposed as a target for a future visit.
Additionally, Rudy requested that we provide a version of SEDRI-LIMS to BD's training centre for internal instrument to LIMS integration training and pledged that BD would notify Arcta of potential instrument installations or upgrades in Africa where SEDRI-LIMS might be appropriate to deploy.
Article first published: 02/06/24