Guide — Labs & Diagnostics
Point-of-Care Testing & Blood Gas Quality Assurance
Bedside analyzers give you a blood gas in minutes instead of waiting on the central lab - but a fast result is only useful if it's a true one. This guide covers the calibration, quality control, and proficiency testing that keep the analyzer honest, and the pre-analytic sampling errors that are the RT's biggest lever on accuracy.
8 min read · Labs & Diagnostics
Written by Apex Respiratory Editorial Team
Educational use only. This material supports respiratory therapy education and exam review. It is not medical advice and is not a substitute for clinical judgment, institutional protocols, or physician orders. Always follow facility policies and current provider orders, and verify calculations independently before clinical use.
Overview
Point-of-care testing (POCT) brings the analyzer to the bedside. Instead of sending a sample to the central lab and waiting, portable blood gas, electrolyte, and co-oximetry analyzers - along with glucose and lactate meters - return a result in minutes. The trade is a small accuracy and scope tradeoff for a large gain in turnaround, which in a critically ill or rapidly changing patient is often the difference that matters.
Respiratory therapists commonly operate and maintain the blood gas analyzers, which means the accuracy of the number is partly the RT’s responsibility. A blood gas drives ventilator changes, oxygen titration, and acid–base decisions, so a result that is wrong because of a bubble, a delay, or a failed control is not a harmless error — it steers the next intervention. Knowing how to keep the analyzer trustworthy is as much a part of the job as drawing the sample.
What POCT covers and who owns it
The point-of-care family is broader than blood gases, but the RT’s lane is well defined:
- Blood gas analyzers— pH, PaCO₂, PaO₂, and calculated HCO₃⁻, run at the bedside on arterial, venous, or capillary samples. This is the device RTs most often operate and maintain.
- Co-oximetry— measured SaO₂ plus dyshemoglobins (carboxyhemoglobin, methemoglobin), which a pulse oximeter cannot distinguish from oxyhemoglobin.
- Electrolyte, glucose, and lactate panels— often bundled into the same cartridge-based analyzer, giving Na⁺, Cl⁻, and other values alongside the gas.
Because the RT runs the device, the RT also runs its quality program: calibration checks, control samples, sensor and membrane changes, and the documentation that proves the analyzer was working when the patient’s result was reported.
Calibration, quality control, and proficiency testing
A trustworthy analyzer rests on three layers of checks, each answering a different question.
Calibration sets the analyzer against known standards. One-point and two-point calibrations - run automatically and on a schedule - anchor the sensors to reference values so the device reads accurately across its range. A drifting or uncalibrated analyzer reports numbers that look plausible but are wrong.
Quality control (QC) proves the analyzer is reading correctly right now. Commercial control samples are run at multiple levels - typically low, normal, and high - and each measured value must fall within its defined range before patient results are reported. When a control falls out of range, you stop: resolve the problem (recalibrate, change reagents or sensors) and re-run the control before resuming patient testing. Reporting a patient result on a failed control is the cardinal QC sin.
Proficiency testingverifies accuracy against the outside world. External, blinded samples are sent in periodically and your results are compared across laboratories, confirming that your analyzer agrees with everyone else’s. In the United States this is a regulatory requirement under CLIA. Maintenance, sensor and membrane changes, and the documentation that ties it all together round out the quality assurance program.
Pre-analytic errors: the RT's biggest lever
Most “wrong” blood gases are not analyzer failures — they are sampling errors that happened before the blood ever reached the device. This is where the RT has the most control, and where a careful technique prevents a false result from driving a bad decision.
| Sampling error | Effect on the result | Prevention |
|---|---|---|
| Air bubbles in the sample | Blood equilibrates toward room air: PaO₂ rises (toward ~150 on room air), PaCO₂ falls. | Expel any bubbles immediately after drawing, before mixing and capping. |
| Excess liquid heparin | Dilutes the sample and, being acidic, lowers the pH and PaCO₂. | Use the minimal heparin in a pre-heparinized syringe. |
| Delay in analysis | Cells keep metabolizing - PaO₂ falls, PaCO₂ rises, pH drops as oxygen is consumed and CO₂ produced. | Analyze within ~15 to 30 minutes, or place the sample on ice. |
| Inadvertent venous sampling | Gives a falsely low PaO₂ and SaO₂ that can mimic hypoxemia. | Confirm an arterial draw; reconcile a low PaO₂ that does not fit the patient. |
A subtler version is “leukocyte larceny”: in marked leukocytosis or thrombocytosis, the elevated cell count accelerates ongoing metabolism in the syringe, so the PaO₂ can fall even faster than usual in a delayed sample. Always label the FiO₂ and the patient’s temperature with the sample, because the interpretation of the gas depends on both.
What the RT does with it
The quality program turns into a short, repeatable bedside routine:
- Run and document QC before reporting. Confirm the controls are in range, and if a control fails, stop testing patients on that analyzer until it is resolved.
- Collect the sample cleanly. Expel air bubbles immediately, use the minimal heparin in a pre-heparinized syringe, keep the draw anaerobic, and mix the sample after capping.
- Analyze promptly or ice. Run within roughly 15 to 30 minutes, or place the sample on ice if analysis will be delayed.
- Record the conditions.Label the FiO₂ and the patient’s temperature so the result can be interpreted correctly.
- Reconcile a result that doesn’t fit. A falsely normal PaO₂ from a bubble, or a falsely low PaO₂ and SaO₂ from an inadvertent venous draw, should be suspected and re-checked before you act on it.
Common Pitfalls
- Reporting on a failed control. If QC is out of range, the analyzer has not earned the right to report a patient result. Resolve it first.
- Ignoring a bubble or a delayed, un-iced sample. A bubble pushes the PaO₂ up toward room air and the PaCO₂ down; a warm delay drives the pH and PaO₂ down and the PaCO₂ up. Both look like real physiology.
- Over-heparinizing. Excess liquid heparin dilutes the sample and, being acidic, lowers the pH and PaCO₂. A pre-heparinized syringe with minimal heparin avoids it.
- Failing to record the FiO₂. Without the FiO₂ and temperature, the reader cannot judge whether the PaO₂ is appropriate or alarming.
- Trusting a number that contradicts the patient. When the gas does not fit the clinical picture, suspect a sampling error before you change therapy.
Board Exam Pearls
- Air bubble raises the PaO₂ and lowers the PaCO₂ - it pulls the sample toward room air.
- Delay without ice lowers the pH and PaO₂ and raises the PaCO₂ as cells keep metabolizing.
- QC must pass before patient results are reported; an out-of-range control stops testing.
- The RT controls most pre-analytic error - technique, not the analyzer, is usually the source of a bad gas.
- POCT trades a small accuracy and scope tradeoff for speed— the value is fast turnaround at the bedside.
FAQ
What does an air bubble do to a blood gas?
An air bubble equilibrates the sample toward room air. It raises the PaO₂ - pulling it toward roughly 150 mmHg on room air - and lowers the PaCO₂. Expel any bubbles immediately after drawing the sample, before mixing and capping, so the blood never has a chance to equilibrate with the air.
Why must I ice a sample that will be delayed?
Blood cells keep metabolizing after the sample is drawn. At room temperature they consume oxygen, so the PaO₂ falls; they produce CO₂, so the PaCO₂ rises; and the pH drops. Analyzing within about 15 to 30 minutes, or placing the sample on ice to slow metabolism, keeps the result close to what was actually circulating in the patient.
What is quality control versus proficiency testing?
Quality control runs commercial control samples of known value - typically low, normal, and high levels - through your own analyzer; the measured values must fall within defined ranges before patient results are reported. Proficiency testing is external: blinded samples are sent in periodically and your results are compared against other labs, verifying ongoing accuracy and satisfying regulatory requirements such as CLIA in the United States.
Why record the FiO₂ and temperature with the sample?
A blood gas can only be interpreted against what the patient was breathing and how warm they were. The expected PaO₂ depends on the FiO₂, and pH, PaCO₂, and PaO₂ all shift with body temperature. Labeling the FiO₂ and the patient's temperature on the sample lets whoever reads the result interpret it correctly instead of guessing.
Related Resources
Sources
- American Association for Respiratory Care. AARC Clinical Practice Guideline: Blood Gas Analysis and Hemoximetry. Respir Care. 2013;58(10):1694-1703.
- Kacmarek RM, Stoller JK, Heuer AJ. Egan's Fundamentals of Respiratory Care. 12th ed. Elsevier; 2021. Blood gas analysis, quality control, and point-of-care testing.