CEO SUMMARY: In Japan, many clinical laboratories are in their third decade of using automation. At the University of Tokyo Hospital, total laboratory automation (TLA) was first implemented in 1991. Now on its fourth generation TLA system, this laboratory was worked upstream to automate specimen collection and urine collection, transport, and specimen preparation. The result is automation solutions not seen in the United States.
JAPAN IS THE UNQUESTIONED WORLD LEADER in clinical laboratory automation. One example of a highly-automated hospital laboratory can be found at University of Tokyo Hospital, in Tokyo, Japan.
Last month, while in Japan, THE DARK REPORT was privileged to tour this laboratory. University of Tokyo Hospital (UTH) has 1,210 beds and its laboratory serves a related outpatient clinic that sees approximately 3,300 patients per day.
The tour was arranged by Sysmex Corporation and was conducted by Hiromitsu Yokota, Ph.D., Technical Supervisor, Department of Clinical Laboratory at UTH. Because the tour was conducted in Japanese and translated to English, THE DARK REPORT acknowledges, in advance, that any inaccuracies in the information which follows are probably due to translation errors.
The high volume core laboratory at UTH is an open lab design. A phlebotomy and specimen collection area is located at one end of this laboratory space. UTH draws approximately 2,000 patients per day and handles 9,000 individual tubes and specimens daily. The majority of patient blood draws are done at this location. Phlebotomy is also done in several other areas of the hospital and tubes are transported to the lab via a pneumatic tube system.
The extent of automation at the UTH laboratory can be illustrated by describing two ways in which phlebotomy is supported by automated systems. First, automation is used extensively to support phlebotomists in the collection process. Patient test requests are transmitted to one of two automated systems that verify the information, then prepare the specimen collection supplies needed for each individual patient. These automated systems prepare bar code labels, pick the right tubes or other collection supplies, apply the labels and then produce a collection tray specifically for that patient.
For the automated system that supports the primary collection site next to the high volume core laboratory, collection trays are sent via the automated line to individual phlebotomy stations. When the collection tray arrives at a phlebotomy station, the phlebotomist then calls the patient and collects the specimens. Without leaving the collection station, the phlebotomist then puts the collection tray on an automated line which transports it directly to the specimen processing area in the main lab.
The second automated phlebotomy system is designed to support phlebotomy being done throughout the hospital. It also receives patient test requests. For each patient, the information is confirmed, then the automated system prints and applies bar code labels to the tubes and other supplies needed for that patient’s collection. A single patient’s collection supplies are then sealed in a plastic bag which is sent, via pneumatic tubes, to the specific site in the hospital where the phlebotomist can retrieve it, perform the collection, then send the pre-labeled specimens back to the lab.
A second noteworthy example of automation in specimen collection at the University of Tokyo Hospital laboratory is a fully-automated urinalysis line. To my knowledge, nothing like this exists in a laboratory in the United States. This urinalysis line was remarkable to watch in operation.
At the main specimen collection center (located adjacent to the core laboratory) exist several urine collection rooms for patients, each with a toilet and sink. Every urine collection room has a pass-through window that opens directly into the laboratory. The patient takes his/her filled paper cup, opens the pass-through door, and places the filled cup of urine directly into the automated transport system.
Urine Specimen Transport
To accommodate this, UTH has designed “hocky puck” carriers on the automated line to hold the cup of urine (which contains approximately six to eight ounces of fluid). Once placed on the automated line, the urine specimen begins its journey. As it reaches the specimen prep station, a pipette extracts the urine specimen and loads it into the analyzer.
What happens next is actually quite entertaining. The specimen cup travels to a station where a robotic arm lifts the cup from its carrier puck. The robotic arm next extends over a vitrous china fixture, resembling a toilet. The arm then dumps the specimen into the fixture and a flush of water moves it into the sewer system. The robotic arm then returns the cup to vertical, moves it over a waste hole, and drops it cleanly into the trash receptacle.
This fully-automated urinalysis system was designed by IDS. According to Dr. Yokota, the original cost for the entire installation, including facility preparation, automation, and analyzers, was in the range of U.S.$2 million. This system handles about 200 urine specimens per day.
Keen Interest In Automation
Several observations will help lab directors and pathologists in other labs understand both the philosophy and the investment in automated support for phlebotomy and a fully-automated urinalysis system as described above.
One, total laboratory automation (TLA) was first installed at the UTH lab in 1990. The lab is now in its fourth generation TLA system. Efforts to automate phlebotomy only came after lots of effort and investment in automating pre-analytical (specimen processing) and analytical stages.
Thus, once the laboratory had realized productivity gains, reduced the number of errors, and achieved less variability because of automation of pre-analytical and analytical stages, lab managers next focused on the remaining largest source of lab errors and variation in work processes. That logically led them upstream of specimen processing, into specimen collection, phlebotomy, and specimen transport to the lab.
Two, automation is viewed as a way to achieve similar improvements in productivity, error reduction, and less variability in collecting and handling specimens before they are received by the laboratory. Automating as much of the specimen collection and transport process as possible is therefore a desirable goal. Further, this automation would allow specimens to flow directly into the pre-analytical stages without human intervention, further serving the ideals of total laboratory automation.
Three, the high number of specimen collections done daily (about 2,000 patients and about 9,000 tubes and other types of specimens) at this laboratory site supported the economics of automation. Such high volume means that return on investment (ROI) can be rapid.
By contrast, few hospital laboratories in the United States come close to drawing 2,000 patients per day from captive inpatient and outpatient populations on their medical campus. That means the ROI calculations for this same phlebotomy automation solution would be much different.
My point here is to call attention to the fact that Japanese hospital laboratories are now almost 30 years into total laboratory automation. Relative to hospital laboratories in the United States, Japanese laboratories support a larger daily volume of work and have much more experience with automation than their counterparts in the United States.
