Flow-Dependent Volume Calibration Failure Before a Pediatric Study

Not acceptable. Volume-accuracy tolerance is conventionally +/-3.5% of the injected volume, so a 3-L syringe must read roughly 2.90-3.10 L. At 2.86 L the slow push is about -4.7% and fails; the fast push at 3.01 L passes. The discrepancy is therefore not a simple gain error.

• A gain error scales all volumes equally, so both flows would read low by the same percentage.
• A leak or partial obstruction is flow- and time-dependent. A slow 3-L push takes much longer, so any leak at the syringe valve, mouthpiece adaptor, or tubing has more time to bleed volume and the slow reading falls further than the fast one. A partially occluded or water-fouled pneumotach can produce the mirror-image pattern, reading low at low flow where the signal is weakest.
• The temperature drop matters because the sensor and the BTPS assumptions are calibrated at a stable ambient condition; a cold sensor can drift, and the running log should note the room temperature.

The pattern points at hardware, not at the patient-facing software.

Sequence matters, because the cheapest explanations come first.

1. Check the syringe itself: valve seating, cracks, plunger seal; a second syringe if available. A faulty reference standard produces a false failure.
2. Check the patient circuit: tubing connections, the mouthpiece adaptor, and the filter; look for condensate or mucus in the pneumotach and dry or replace it.
3. Check the environment and settings: temperature sensor reading versus a room thermometer, BTPS correction enabled, and adequate warm-up time after the overnight cold soak.
4. Repeat the two-flow check. It must pass at both flows before any patient is tested.

If it still fails, take the device out of service, tag it, notify the lab director, and move the child to a backup spirometer or another lab.

The temptation is to test the child anyway because the next slot is 3 weeks out and the fast-flow check passed. Resist it: a bronchodilator study compares two numbers with a difference of roughly 100-150 mL, and a flow-dependent error of this size can manufacture or erase that difference. A documented reschedule is a defensible clinical delay; a fabricated reversibility result is not. Note the reschedule reason and escalate the slot if the child's care depends on it.

A biological control tests the whole chain - device, circuit, and a real forced maneuver - and is judged against that person's established baseline, commonly within about 5% (roughly 0.2 L) of it.

• Control also reads low (FEV1 well under 3.80 L, FVC down proportionally): the device and circuit are at fault, corroborating the syringe finding. Keep the spirometer offline even if a later syringe check happens to pass, and do not test patients until both checks agree.
• Control is within 5% of baseline while the syringe check fails: suspect the syringe or the calibration technique rather than the spirometer, and re-verify with a second syringe before condemning the device.
• Control is low only because the technologist is having a bad day: a single out-of-range control is a prompt to repeat the control, not proof of device failure - but it can never be used to override a failed volume calibration.

Either way the biological control is a supplement, not a substitute: the volume check is the required daily gate, and it has already failed here.