3.3 ABG & Gas Analyzer Troubleshooting and QC
Key Takeaways
- Common blood gas faults include electrode drift, air bubbles, protein contamination, empty calibration materials, and obstructed or leaky sample paths—each has a recognizable pattern.
- Gas-analyzer faults often present as failed zero/span, wet sensors, exhausted dryers/absorbers, or room-air leaks that dilute sample gases.
- Routine QC uses multilevel control materials on a schedule to verify ongoing performance; proficiency testing (external) verifies the lab against peer/assigned values (DCO I.C.2).
- Runs that fail Westgard-style or manufacturer rules, or that follow failed calibration, are rejected; repeated failure requires taking the analyzer out of service until corrective action and re-verification succeed (I.B.2, I.B.7, I.C.7).
- Document corrective actions: what failed, what was replaced or recalibrated, which QC passed, and when patient testing resumed.
Troubleshooting Is a Domain I Skill
The PFT Detailed Content Outline separates I.B Troubleshoot from I.C Quality Control, but at the bench they form one loop: detect a problem → isolate the subsystem → correct it → verify with calibration/QC → release or continue holding results. For RPFT high-cut performance, pattern recognition beats random part-swapping.
Relevant outline anchors for this section include troubleshooting blood gas and gas systems (I.B.2, I.B.7 in the equipment-troubleshooting cluster) and QC/proficiency activities (I.C.2, I.C.7). Exact lettering on a printed DCO should be checked against the current NBRC document; the competencies—recognize failure modes, apply QC rules, and stop unsafe testing—are stable.
Blood Gas Analyzer: High-Yield Fault Patterns
| Symptom pattern | Likely causes | First actions |
|---|---|---|
| Gradual drift of one analyte across shifts | Electrode aging, membrane wear, protein film, temperature instability | Inspect/clean path; membrane or electrode service; two-point cal; multilevel QC |
| Sudden PCO2 ↓ and PO2 ↑ toward room-air values | Air bubbles in syringe or analyzer path; loose fitting; incomplete aspiration | Reject sample; recollect without bubbles; purge/prime path; check seals |
| Slow equilibration / "busy" flags | Protein contamination, clot, depleted electrolyte, cold module | Rinse/enzymatic clean per SOP; replace membrane/electrode; confirm 37 °C |
| Calibration fails on one channel only | Empty/expired cal gas or buffer for that channel; bad electrode; cartridge fault | Replace consumable; service channel; do not report that analyte |
| All channels fail after pack change | Pack not seated, wrong pack, air in reagents, incomplete prime | Reseat/replace pack; prime; full cal + QC before patients |
| Carryover (second sample biased by first) | Inadequate rinse, short path flush, clot trapping blood | Extend rinse; clear obstruction; reanalyze after clean blank/flush |
Electrode Drift
Drift is slow loss of calibration integrity. One-point calibrations may mask mild drift until QC fails at a level far from the calibration point (slope error). When controls show a trend—especially the same direction on consecutive runs—intervene before a hard failure. Trend review is part of quality management, not optional chart decoration.
Air Bubbles
Air bubbles are both a preanalytical and analyzer-path problem. Room air has high PO2 and near-zero PCO2 relative to venous or many arterial samples:
- PO2 is pulled toward ambient,
- PCO2 is pulled down,
- pH may rise as PCO2 falls.
If the bubble is in the syringe, recollect or carefully expel air immediately per policy (timing and technique matter). If the bubble is in the instrument, stop and purge; do not average a bubbled reading with a clean one and call it science.
Protein Contamination and Clots
Blood is sticky. Protein and fibrin coat membranes and narrow capillaries. Response time lengthens first; gross bias follows. Prevention: proper anticoagulation of the specimen, prompt analysis, adequate rinses, and scheduled deep cleans. A clot that occludes the path can cause partial sampling—another reason "the number looks odd" may be mechanical, not physiologic.
Empty Calibration Materials
Empty cal gas tanks, depleted onboard generators, or dry buffer bottles produce hard calibration failures. The correct mental model: no valid calibration → no patient reporting for affected analytes. Swapping to a backup analyzer is preferable to creative interpretation of failed cal flags.
Gas Analyzer and Absorber Fault Patterns
| Symptom | Likely causes | First actions |
|---|---|---|
| Zero OK, span fails | Wrong span cylinder, empty cylinder, leak between cal port and analyzer, dying IR source/TCD | Verify gas certificate and connections; leak check; service analyzer |
| Span OK on cal port, patient samples diluted | Leak in sample line, cracked Nafion, loose valve block | Pressure/leak test the patient path; replace dryer tubing |
| Unstable He baseline | Moisture on TCD, flow instability, temperature swing | Dry the train; stabilize flow; re-zero after warm-up |
| Wet IR windows / slow DLCO CO response | Exhausted Drierite, failed Nafion purge, condensation | Replace desiccant; service dryer; dry and recalibrate |
| Progressive failure after many tests | Absorber exhaustion, dirty filters, aging fuel cell O2 sensor | Replace consumables on schedule, not after catastrophe |
Leaky sample lines deserve special emphasis. A spirometer leak and a gas-analyzer leak both destroy data, but gas leaks often dilute tracer and CO, producing DLCO or dilution-volume errors that look like patient disease. Always leak-check the gas path after maintenance.
Routine QC vs Proficiency Testing (I.C.2)
Routine (internal) QC
Routine QC uses commercial or manufacturer control materials at multiple levels (commonly low, normal, high for blood gases) on a defined schedule—often each day of use or each shift, plus after major maintenance. Controls verify that the currently calibrated system recovers known values within statistical or manufacturer limits.
Core practices:
- Run the levels required by SOP and regulatory/accreditation standards applicable to your lab,
- Plot or review results for shifts and trends (Westgard-style rules are the conceptual language even when software automates them),
- Use the same sample path patients use when the method requires it,
- Document lot numbers and expiration dates.
For gas analyzers, "QC" may include periodic verification with certified gases, biological controls (healthy subject repeatability for spirometry/DLCO systems), and mechanical checks—still the idea of comparing observed versus expected under controlled conditions.
Proficiency testing (external)
Proficiency testing (PT) ships unknown or blinded samples from an external provider. The lab analyzes them like patient samples and returns results for scoring against assigned values or peer groups. PT evaluates the total testing process under real conditions and is not satisfied by merely running internal QC the same morning.
| Feature | Routine QC | Proficiency testing |
|---|---|---|
| Source of material | Lab-purchased controls | External PT provider |
| Frequency | Daily/shift/per SOP | Scheduled events (e.g., multiple times per year) |
| Purpose | Continuous readiness | External verification / accreditation evidence |
| When failed | Investigate, correct, possibly stop testing | Investigate, correct, report per PT program rules; may trigger broader review |
Exam trap: treating PT failure as "just a bad ampule" without investigation, or claiming internal QC alone replaces PT.
Rules for Rejecting Runs and Taking Analyzers Out of Service (I.C.7)
Rejection is a professional duty, not a customer-service failure.
Reject (do not report) when:
- Calibration failed or was incomplete for that analyte,
- QC is out of limits and not resolved by documented, allowed corrective action,
- Sample integrity is compromised (clot, bubble, wrong syringe, unlabeled specimen—preanalytical rules still apply),
- Instrument flags indicate incomplete aspiration, temperature fault, or air in path,
- PT or internal surveillance shows the method is not under control.
Take the analyzer (or channel) out of service when:
- Repeated QC failures persist after recalibration and basic maintenance,
- Critical hardware faults occur (temperature control loss, persistent aspiration failure, optical failure on co-ox),
- Required reagents/packs are unavailable or expired with no validated alternative,
- Corrective action cannot be verified before results are needed.
Out-of-service steps should be standardized:
- Stop patient testing on the affected channel/device.
- Label the instrument (physical tag + LIS/middleware block if available).
- Perform corrective maintenance or call service.
- Calibrate and run multilevel QC (and any required linearity or verification).
- Document return-to-service with date, time, tech ID, and QC data.
- Resume testing only after all required levels pass.
Scenario. Morning QC: Level 2 PO2 is 1.5 SD low; Level 3 PO2 fails low beyond limits; pH and PCO2 pass. The technologist recalibrates PO2; Level 3 still fails. Correct path: hold PO2 reporting, replace PO2 membrane/electrode or cartridge per SOP, recalibrate, pass all PO2 QC levels, then release. Incorrect path: report pH/PCO2 only while "estimating" PO2 from SpO2, or average the failed control with the previous hour’s pass.
Integrating Troubleshooting with QC Documentation
Every failure event should leave a trail that answers four questions: What was wrong? What fixed it? How did you prove it? When did patient testing resume? That documentation supports accreditation, protects patients, and is the operational expression of Domain I.C competencies.
Study Hook
An arterial sample analyzed on a blood gas instrument shows unexpectedly high PO2 and low PCO2 compared with the patient’s clinical status and recent results. Multiple small air bubbles are visible in the syringe. What is the best interpretation?
How does proficiency testing differ from routine daily QC for a blood gas analyzer?
After electrode service, two-point calibration passes, but multilevel PO2 QC fails limits twice despite repeating the controls carefully. What should the technologist do?
A PFT gas analyzer spans correctly when calibration gas is applied at the analyzer inlet, but DLCO patient samples read as if diluted. Drierite indicator is fully spent and Nafion tubing is kinked. The most appropriate first priority is to: