Quality Assurance

Laboratories subscribe to external programmes that post unknown tissue sections or blocks several times a year. The technologist stains or processes that material and returns results, which are then graded against peer laboratories. The point is only served if the challenge travels the ordinary route: the staff who work the bench, the instruments in daily use, and the written method already in force. Seeking the answer from another laboratory defeats the exercise. A graded score that comes back unacceptable is not simply filed; it opens an investigation into why, and the fix is recorded.

When something goes wrong, whether a missed reagent change, a mislabelled cassette, a processor fault or a step skipped under pressure, the technologist is expected to say so through the laboratory's incident system rather than quietly correcting it. The report names what happened, which cases were touched, and what was done immediately to limit harm. Reports are not disciplinary tools; they are how repeat faults become visible and how systems get fixed. Concealing a deviation removes the only evidence that a patient's blocks may have been compromised, and it makes the same failure certain to recur.

Much of histology quality work lives on paper or in a spreadsheet: flotation bath and oven temperatures, cryostat chamber readings, processor solvent rotations, microtome blade changes, fridge and freezer monitoring, stain lot numbers, and charts that plot those numbers over time. The record must be made at the time by the person who did the check, with an initial and a date. Values outside limits need a note saying what was done. Corrections are made so the original entry stays readable, never erased or painted over, because an unreadable record is treated as no record at all.

Every special stain and every immunohistochemistry run carries controls, and the technologist reads them before releasing the patient slide. A positive control confirms the reagent chain worked; a negative reagent control confirms that what is seen is genuine binding rather than background. Instrument calibration data are judged the same way, against limits set in advance. Numbers plotted over time also carry meaning: a run of points sitting on one side of the mean signals a systematic change, while points climbing steadily suggest gradual drift. Failing controls invalidate the run, and no patient result leaves the laboratory from it.

A CD20 immunohistochemistry run finishes and the technologist checks the controls. The negative reagent control section, shown here, carries diffuse brown chromogen across the tissue, and the patient section on the same slide rack is broadly brown as well. What does this tell the technologist about the run?

APatient positivity is confirmed by both slidesBChromogen development was slightly too shortCThe primary antibody titre needs increasingDThe run is invalid and has to be repeated

Daily cryostat chamber temperature readings are plotted on the chart below, with the established mean at −25.0 °C and one standard deviation equal to 1.0 °C. The first eight days scatter closely around the mean; the last six sit together at roughly −23.5 °C without climbing further. How should this pattern be described?

AA trend caused by gradual reagent ageingBA shift caused by a systematic changeCRandom error inherent to the measurementDAn outlier that should be repeated once

Nothing new goes straight onto patient tissue. A fresh lot of eosin, a new antibody clone, a different paraffin, a replacement processor or a newly serviced microtome all have to demonstrate that they perform at least as well as what they replace. In practice this means staining control tissue of known behaviour side by side with the material in current use and comparing the sections under the microscope. Antibodies need tissues expected to stain and tissues expected not to. The comparison, the operator and the date are recorded, and only then does the item enter service.

Inventory work keeps the bench running without waste. It covers knowing what is on the shelf, ordering before stock runs out, receiving and checking deliveries against the order, recording lot numbers and expiry dates, and storing items as the supplier directs. Stock is rotated so the oldest usable container is drawn down first and newer stock sits behind it, which keeps material from quietly reaching expiry at the back of a cupboard. Expired reagents are removed from the working area rather than left available. Minimum stock levels for critical items are set so a routine order never becomes an emergency.

Histology generates identifiable data at every step: requisitions, accession numbers, block and slide labels, scanned images, and the reports themselves. The technologist is responsible for entering it accurately, keeping the link between patient, specimen, block and slide unbroken, and releasing it only through approved routes to people entitled to receive it. Whole-slide images and photographs count as patient records even when a name does not appear on the file. Personal email, removable drives and shared logins fall outside the laboratory's controlled paths, so requests that would use them are redirected rather than accommodated, however senior the requester.

This is the histology-specific piece of quality work: looking at the finished slide and deciding whether a pathologist can safely read it. The technologist checks nuclear and cytoplasmic balance, differentiation, the colour a special stain has produced against its control, and whether processing or cutting has left a defect such as folds, chatter, knife lines or an out-of-focus scan. Faults affecting a whole rack usually point at a bath or reagent; faults on one slide point at the block or the section. An unacceptable slide is recut and restained rather than sent out with a caveat.

A full rack of routine sections comes off the automated stainer and every slide looks the same under the microscope: nuclei are faint and washed out while cytoplasmic and connective tissue staining is of normal intensity. Which explanation fits this finding?

AThe haematoxylin bath is exhaustedBThe sections were cut too thin at 2 µmCFixation in formalin ran past 48 hoursDThe bluing reagent was left in too long

A Grocott methenamine silver on a lung biopsy is checked at the microscope before release. Dense black granular deposit lies over the whole section, as shown, obscuring tissue detail, and the fungal control stained in the same run is equally dark. What should the technologist do with it?

ARelease it with a note about the dark backgroundBRecut and restain with fresh silver solutionCDecolourise the slide and add a light green rinseDRead only the areas where deposit looks thinnest

Every method in the laboratory exists as a controlled document carrying a version number, an effective date and evidence of review. The technologist works from the current version only, and knows where to find it. Personal shortcuts, photocopied pages taped to cupboards and annotated printouts sit outside that system, and once a procedure is revised those copies quietly become wrong. Changes are made through the review process rather than by editing a bench copy. Obsolete versions are withdrawn from circulation, and staff sign to show they have read a revision before it takes effect.