GEM 5000 vs Lab Sysmex XN Hemoglobin

21 followers

0 Likes

I'm just trying to help out our respiratory department by inquiring about this topic. They use the GEM 5000 instruments and are doing semi-annual comparisons between them and our lab instrumentation. We are seeing fairly large differences in the Hemoglobin results between the GEMs and the Sysmex. GEM 5000 proficiency samples look great for Hemoglobin, as well as the Sysmex proficiency Hemoglobin results, they just don't compare well to each other. What are other facilities doing about this? What % difference are you using as acceptable for the Hemoglobin results between lab and blood gas instruments? Are you using a certain sample for this? They are using patients with lines and drawing a syringe for their GEM 5000 instrument and then an EDTA tube to run on the Sysmex. Again, just trying to see what others are doing and forwarding this information along. Any input is appreciated. Thanks!

6 Replies

Jo Win

Mon Mar 02, 2026 at 3:03PM

The discrepancy may be due to pre analytical errors.
Are they flushing the line for any contaminants?
Are the specimens well mixed when testing?
Is the GEM calculating or doing a direct measurement for the hemoglobin? GEMs can be setup to calculate or directly measured. How was your's setup?
If the GEM is calculating the hemoglobin while the Sysmex directly measures, you will see discrepancies.

Remember that you can't do method correlations with a calculation.

James Beck

Mon Mar 02, 2026 at 3:23PM

GEM 5000 has a cooximeter in it, so total hemoglobin is directly measured, not calculated. That said, they should be comparing fairly well. How much difference are you seeing? Also, is there a low or high bias towards one method or the other? If you have a bias, you may be able to add an adjustment to the slope of one method to match better. If no bias, and if the difference is within the expected SD/CV of either machine, you may not have a problem.

Ashlee Holland

Mon Mar 02, 2026 at 3:37PM

James - the difference isn't with all of the comparison samples but for example:
GEM #1 tHb 12.9 GEM # 2 tHb 12.9 GEM # 3 tHb 13.0 Sysmex Hgb 12.0
GEM #1 tHb 12.8 GEM # 2 tHb 13.3 GEM # 3 tHb 12.9 Sysmex Hgb 10.0
GEM #1 tHb 8.7 GEM # 2 tHb 8.9 GEM # 3 tHb 8.7 Sysmex Hgb 7.3
GEM #1 tHb 11.0 GEM # 2 tHb 11.6 GEM # 3 tHb 10.9 Sysmex Hgb 7.5
GEM #1 tHb 8.10 GEM # 2 tHb 8.50 GEM # 3 tHb 8.50 Sysmex Hgb 6.8

Some are a little concerning in the differences, and others not so much

James Beck

Mon Mar 02, 2026 at 3:46PM

yes, samples 2 and 4 in particular. It does appear that the GEM results are trending higher. THb values that trend higher in blood gas samples can often be attributed to undermixed samples, where the sampling probe aspirates a hemoconcentrated part of the sample and makes the tHb values trend higher - whereas automated cell counters have an automated mixing step by default. I would start there.

Miranda Cotter

Wed Mar 04, 2026 at 11:21AM

We run into this occasionally, especially when the hemoglobin is on the lower end. For correlating our GEMs to our Sysmex XN, we use 7% or 0.5 g/dL for acceptable criteria.

Pamela Knaack

Thu Mar 05, 2026 at 9:09AM

We’ve had similar discussions at our site, and one thing to keep in mind is that there really is no universal or “required” % difference for hemoglobin correlation between blood gas instruments and central lab analyzers. Correlation limits can be whatever makes sense for your institution, as long as they are data‑driven and approved by your medical director(s).

What many facilities do (and what we’ve done) is:

Perform their own correlation studies between the GEM (or other blood gas analyzer) and the hematology analyzer using their patient population and workflows.
Look at bias, trends, and clinical impact, not just a single % difference.
Establish site‑specific acceptance criteria based on those findings rather than trying to match what another hospital uses.
Document the rationale and have the lab and POC medical directors sign off. Once that’s done, you’re generally in good shape from a regulatory standpoint.

It’s also important to remember that:

Blood gas hemoglobin and Sysmex hemoglobin are methodologically different (whole blood co‑oximetry vs. impedance/optical methods), so some systematic bias is expected.
Excellent proficiency testing performance on both systems does not guarantee tight agreement between the two methods.
Pre‑analytical factors (line draws, dilution, anticoagulant differences, timing between samples) can significantly affect comparisons, even when best practices are followed.

Because of all this, what another site considers “acceptable” may not be appropriate for yours. Their limits are often specific to their instruments, patient mix, and clinical expectations.

Bottom line: do your own correlations, decide what differences are clinically acceptable for your providers, document it well, and get medical director approval. Other sites’ practices can be helpful for context, but they shouldn’t drive your acceptance criteria.

Subgroup Membership is required to post Replies

Join POCT Listserv now

Topic	Replies	Views	Last Reply
Blood Gas/Chem POC devices Kristine Zinn 1 day ago	1	385	Jeremy MacDonald about 23 hours ago
Epic-Beaker Venous Draw Requests for Unsolicited Orders James Beck 1 day ago	3	360	Barbara Edmondson 1 day ago
Blood Gas syringe Heparin concentration Jessica Jenkins 1 day ago	0	226	Jessica Jenkins 1 day ago

Setup Information

GEM 5000 vs Lab Sysmex XN Hemoglobin