Bring point-of-care testing to your ED

Many progressive emergency departments (EDs) are using point-of-care testing to get test results in seconds at the patient’s bedside, reports Beverly Giles, RN, an emergency nurse at Methodist Hospital in Indianapolis. "We’ve seen this expand in recent years, and it will continue to evolve," she says. "However, you need to examine all the pros and cons before deciding to do point-of-care testing in your ED."

Here are some pros and cons of point-of-care testing to consider:


Fast results. At Central Peninsula, point-of-care testing has speeded results for several lab tests. "Before, we could wait up to two hours to get STAT results, now we can get a turnaround of 90 seconds for glucose, hematocrit, hemoglobin, and electrolytes, and 120 seconds for blood gas analysis," reports Johnson.

A small amount of the patient’s blood is drawn and injected into a thin cartridge. The cartridge is then inserted into the hand-held analyzer. "You put it into the machine, and by the time you are finished entering your operator number and the data into the medical record, it’s about ready to give you the results," says Johnson.

Methodist’s ED compared lab times before and after point-of-care testing was implemented. "Before we decided to implement i-STAT housewide, our lab did a study and figured the average turnaround time for a Chem 7 was 20 minutes. With i-STAT, it went down to 5 minutes, including all the steps of drawing the blood," Giles reports.

However, waiting for results are only one component of patient delays, notes Giles. "It may not have a big effect on quicker dispositions, because in the ED usually lab work is not the only thing you are waiting for. If the patient is going to be admitted, there is also the issue of a bed, consults, or a CAT scan."

Patients can be treated more quickly. "It’s a benefit to make medical decisions more quickly right there at the bedside. Rather than waiting for the lab to run the test and notify you of the results, it’s right there in front of you," says Giles. "Patients with arrhythmia, on ventilators, or multiple trauma patients who need blood, can be treated more quickly."

Delays in treatment are reduced. "Any time you have a change in chloride, potassium, or sodium you can certainly start to treat a lot faster if you know those results," says Johnson. "If some fine tuning is pending on results of those electrolytes, it’s easier to monitor a patient who may be losing blood such as a GI bleed, because you can measure that hematocrit. It’s also a lot easier to get small sticks at bedside versus having the lab come over and draw it all the time."

The local EMS system worked with the vendor to put the system into ambulances. "This way, they can roll in and give us the results as they come through, which certainly speeds things up," says Johnson. "If the ambulance is five minutes away from your facility it won’t help you that much, but in rural Alaska it sometimes takes over one hour for them to reach the hospital, so it makes a big difference."

Prehospital providers can get a better sense of internal injuries of trauma patients (See sidebar on prehospital use of point-of-care testing.) LifeGuard Alaska, a medevac team from Anchorage, AK also uses i-STAT. "In Alaska, some medevacs are over four hours away, so these results certainly help treatment plans. They can keep an eye on the hematocrit to determine if the patient is losing blood, which will help us be better prepared," Johnson explains. "We can have surgery on standby if needed. This way, we know when they hit the door where our focus should be."

Testing is facilitated for some patients. "It only takes two drops of blood to do all this testing. It’s so great when have limited venous access with a patient," says Giles.


Potential for decreased accuracy. When ED nurses at Central Peninsula General Hospital in Soldotna, AK, became dissatisfied with accuracy of the point-of-care testing system being used, a more accurate system was requested. "We had been using the One Touch glucose monitor, and people weren’t very good throughout the facility about doing controls on it. We were finding that they weren’t as accurate as we needed them to be consistently," says Jan Johnson, RN.

The One Touch system was dependent on operator technique to yield accurate results, says Johnson. "Small variations in sample volume led to large variation of results," she explains. "In addition, the liquid controls were not being run as required, and matching the calibration value to the specific lot of reagent strips was often overlooked."

Several point-of-care products were evaluated based on the following criteria, says Johnson: accuracy, precision, portability, ability to identify patient and operator ID for each test result, streamlined QC, reduced dependency on operator technique, range of tests available, centralized management of data, ability to integrate with laboratory and nursing routines, and expectation of future support and expansion of available tests. A point-of-care testing device from i-STAT corporation was selected.

One study found looking at point-of-care testing in the ED found that hematocrit results were falsely high.1 "They retrained staff, but still had falsely high results," notes Giles.

At Methodist, some quality assurance issues had to be resolved. "At one point, a cardiologists went to check the hematocrit lab results of one of his patients, and noticed discrepancies between ED i-STAT and the central labs," Giles recalls. "We started to investigate why this was happening, and it became a training issue. Our staff was generally not mixing samples well, and using samples which were longer than 10 minutes old rather than sticking the patient again."

Additional training was done, and only nurses were allowed to do the i-STAT testing. "We did retesting to confirm that results were accurate, and each person had to do several tests that were verified by the lab," says Giles. "Also, if we have to do a fresh specimen, we mix it thoroughly, and do it right at the patient’s bedside."

Also, a decision was made to use i-STAT on a limited basis. "We only use it if we have a physician’s order, instead of anticipating the need," says Giles. "Only certain conditions need lab work immediately, such as diabetics or multiple traumas, not the average run-of-the-mill patient who needs a Chem 7."

Staff need additional training. At Central Peninsula, initial training takes about two hours, using a checklist guide cooperatively developed by nursing services and the laboratory. "Points covered run the range from mechanics of the system to sampling to clinical use of results, and hands-on practice with blood samples," says Johnson. "Annual training is included within the skills update program of nursing services."

Regulations must be adhered to. "The i-STAT system is regulated under CLIA [Clinical Laboratory Improvement Act] as a moderately complex testing system, says Johnson. "As this was the first system installed in Alaska, state compliance inspectors worked with the company to be sure the system was acceptable," she notes.

Patient turnaround times may be decreased. "We started it on a trial basis and found it was the greatest thing to come along," says Giles. "At first we thought it would take up a lot of extra time, but we saw that patients got out faster."

For two months, an audit of turnaround times was done. "We decreased turnaround times by 11 and 15 minutes as compared to previous months," Giles reports. "Multiply that by the 250 patients we see a day, [and it] adds up to a significant savings in personnel time. It’s also good for patient satisfaction."

Staff may view it as extra work. "When we first implemented this, staff felt we are already doing more with less, and now you want us to do the lab work too?’" Giles recalls. "That was the perception, but now it’s second nature."

It may create controversy. The lab department may have strong feelings against point-of-care testing done in the ED. "Often, there is controversy about whether this is needed. A lot of that resistance is due to lab technicians being afraid it will impact their jobs," says Giles.

At Central Peninsula, the lab was involved in the process. "For this to be a successful venture, we needed the lab’s buy-in," says Johnson. "Initially, there was a little bit of hesitancy because this does take work away from them. We compromised with the ED, getting the revenue for the blood draw fee and the lab getting paid for the actual test, which created a win-win situation."

The lab manager wanted reassurance that studies would be accurate. "We ran concurrent tests for the first few months to assure test-result accuracy," says Johnson."

A correlation study was completed comparing i-STAT results to laboratory results for each analyte before the system was accepted for use. "Continued concurrent testing has been done on a random basis by performing equivalent lab tests on an i-STAT patient and comparing results," Johnson explains. "These comparisons demonstrate that the i-STAT results continue to meet the original correlation standards."

It may be difficult to do the necessary quality control in the ED. "Labs are concerned about accuracy, and run quality controls on instruments a couple of times each shift, and nurses aren’t used to that. That is the mindset of the lab, and nurses are more interested in benefits to patient care," says Giles. "It is a big learning curve for nursing staff to suddenly take this role."

i-STAT is no longer used for glucose testing, which eliminates quality control procedures. "Now we do venous sticks, so patients aren’t as happy with that, but we were able to do away with quality control procedures with the glucometer every day," says Giles. "We have a very large ED, so it was a big issue for us to do the simulator testing every six hours, with the number of i-STATs we had in the department."

It may or may not be cost effective. i-STAT units cost approximately $4500. Whether or not point-of-care testing is cost effective depends on your ED, says Giles. "It depends on how you are set up. The cartridges are fairly expensive, but we saved money because we were able to close the stat labs, so the lab was able to shave a lot off their budget."


To learn more about point of care testing, contact the following:

• Diana Herr, RN, Mayo Medical Center, 1216 2nd Street SW, Rochester MN 55902. Telephone: (507) 255-7233. E-mail:

• Janet Johnson, RN, BSHA, CEN, CFRN, Central Peninsula General Hospital, 250 Hospital Place, Soldotna, AK 99669. Telephone: (907) 262-8154. Fax: (907) 262-0799.

• Beverly Giles, RN, emergency medicine research nurse, Methodist Hospital, 1701 North Senate Blvd., Indianapolis IN 46202. Telephone: (317) 929-3898. fax: 317 929 2306. E-mail:

• i-Stat Corporation, 303 College Road East, Princeton, NJ 08540. Telephone: (800) 827-7828. Fax: (609) 243-9311.


1. Parvin CA. Impact of point-of-care testing in patients length of stay in a large ED. Clin Chem 1996;42:711-717.