{"slug":"sterility-test-positive-investigation","title":"Sterility Test Positive: How to Investigate Without Making It Worse","excerpt":"A sterility positive is one of the most consequential findings in pharmaceutical manufacturing, and how it is handled determines whether the outcome is a contained deviation or a regulatory crisis. When invalidation is legitimate, what the organism tells you, and the patterns inspectors look for.","date":"2026-07-15","author":"DSTechnoverse","image":"/blog-images/microbiology/contamination-investigation-tree.svg","category":"Sterility Testing","tags":["sterility test positive","sterility failure investigation","test invalidation","aseptic processing deviation","batch rejection","media fill failure","GMP investigation"],"content":"A positive sterility test triggers one of the most heavily scrutinised investigations in pharmaceutical manufacturing. The compendia permit invalidation only where it can be demonstrated that growth resulted from a fault in the test itself — not from a general belief that laboratory contamination is likely.\n\n![Structured contamination investigation path](/blog-images/microbiology/contamination-investigation-tree.svg)\n\n## The Governing Principle\n\n**A sterility test may not be invalidated without documented evidence of a specific, identified fault in the test.**\n\n\"Probable laboratory contamination\" is not evidence. \"The analyst may have touched something\" is not evidence. A confirmed media sterility failure, a documented technique deviation recorded at the time, a positive filtration control, or a demonstrated environmental breach during the session — these are evidence.\n\nThe reason the bar is set so high is statistical. The sterility test is a weak instrument: it detects contamination reliably only when contamination is widespread. A positive result is therefore *unlikely to be a false alarm* — if anything, the test is far more likely to miss real contamination than to invent it. Treating positives as probable artefacts inverts the actual probabilities.\n\n## Start by Preserving Everything\n\nBefore anything else, secure the evidence. Investigations are routinely crippled on day one by a discarded plate.\n\n- The isolate, subcultured to purity and cryopreserved.\n- The remaining sample and any untested units from the same batch.\n- Media lot records, growth promotion and sterility test results for that lot.\n- All controls from the session — negative, filtration and positive.\n- Environmental monitoring plates from the test session.\n- The analyst's records and the session log.\n\n## Phase One: Was It the Test?\n\n1. **Identify the organism to species.** This is the most informative single step and everything downstream depends on it.\n2. **Interview the analyst promptly**, using a structured checklist while the session is fresh. Ask what was done, not whether anything went wrong.\n3. **Review media records** — growth promotion for that lot, sterility of uninoculated controls, expiry, storage.\n4. **Review the controls.** A contaminated negative or filtration control is direct evidence of a test fault. Valid controls substantially weaken the laboratory-error hypothesis.\n5. **Review environmental monitoring** of the test session, including operator glove prints and the Grade A record.\n6. **Check the isolator or LAF integrity** — glove integrity testing, pressure records, HEPA integrity, any breach during the session.\n7. **Review incubator records** for the full period.\n8. **Review the analyst's recent history.** Positives clustering around one analyst suggest technique; positives distributed across analysts suggest something else.\n9. **Conclude explicitly.** Either a specific fault is identified and documented, or it is not.\n\n## What the Organism Tells You\n\nThe identity frequently settles the phase one question on its own.\n\n| Organism | Interpretation |\n|---|---|\n| *Staphylococcus epidermidis*, *Micrococcus*, *Corynebacterium* | Human skin flora — could be analyst or production personnel |\n| *Bacillus* and other spore formers | Environment or a sterilisation failure — rarely a laboratory artefact |\n| *Pseudomonas*, *Burkholderia*, *Ralstonia* | Water — laboratory water or process water, separable by strain typing |\n| *Aspergillus*, *Penicillium* | Airborne mould, laboratory or production environment |\n| *Escherichia coli*, *Salmonella* | Should never appear; treat as real without exception |\n| *Propionibacterium*, anaerobes | Skin or a genuine anaerobic process source |\n\nThe critical inference: a common skin organism recovered alongside a failed operator glove print in the same session is a plausible laboratory contamination case. **The same organism with all controls valid is not.** A water organism or an enteric should never be attributed to the laboratory without exceptional evidence.\n\nA spore former deserves particular caution. Spores are what survive a marginal sterilisation cycle, so recovering one is more consistent with a process failure than a laboratory slip, and treating it as an artefact is the error most likely to end badly.\n\n## Phase Two: The Manufacturing Investigation\n\nIf phase one identifies no test fault, the result stands and the investigation extends.\n\n- **Batch record review** — deviations, interventions, hold times, unusual events during the production window.\n- **Sterilisation records** — cycle parameters, load configuration, biological and chemical indicators, any deviation.\n- **Filter integrity** — pre-use and, critically, post-use test results.\n- **Environmental monitoring** for the production area across the period, looking for excursions or adverse trends that coincide.\n- **Media fill history**, including any recent failure or any intervention not covered by simulation.\n- **Water system results** for the period.\n- **Personnel** — who was working, gowning records, glove print trends by individual.\n- **Container closure integrity** data.\n- **Maintenance activity** in or near the area, which correlates with contamination events more often than most facilities expect.\n\nStrain typing against the [facility flora library](/blog/facility-flora-library-microbiology) converts circumstantial correlation into evidence.\n\n## Impact Assessment\n\nScope is determined by what the batch shared with others.\n\n1. Which batches were in the same sterilisation load or campaign?\n2. Which used the same filter, the same bulk, the same components?\n3. Which ran in the same area during the same period?\n4. Which shared personnel or shift?\n5. Is any affected product distributed?\n6. What is the health risk for the patient population, given the organism?\n\nFor a sterile product the last question is rarely academic. An organism in a parenteral reaches the bloodstream directly, which is why the disposition bar is set where it is.\n\n## Disposition\n\nThe decision must be made by quality, documented with a scientific rationale, and must follow from the investigation rather than precede it.\n\nWhere no test fault is established, the batch fails. Release following a genuine sterility positive requires substantially more justification than rejection and is examined closely — in practice it is rarely defensible.\n\n## The Retest Question\n\nRetesting after a sterility positive is where the most serious data integrity findings originate, so the rules deserve stating precisely.\n\n**A retest is legitimate** only when phase one identified a specific fault in the original test. The retest then replaces an invalid result with a valid one. The original result, the identified fault and the retest are all documented, and the retest is performed on retained material by a different analyst where possible.\n\n**A retest is not legitimate** when it is performed because the first result was unwelcome, when phase one found no assignable cause, or when additional units are tested and the original result is quietly averaged away or disregarded.\n\n**Resampling** — taking fresh units from the batch — is more restricted still. It is defensible only where the original sample is demonstrably unrepresentative, for instance where a documented sampling error occurred. Resampling because the first sample failed is not defensible under any framing.\n\nInspectors examine the ratio of retests on failing results to retests on passing results. A laboratory that retests only failures has revealed its decision rule regardless of what the procedure says, and that pattern is visible in the data long before anyone admits to it.\n\n## The Patterns Inspectors Look For\n\n- A high proportion of positives invalidated as laboratory error, especially without identified specific causes.\n- Retests performed before phase one concluded.\n- Investigations closed implausibly quickly.\n- Conclusions not following from the evidence presented.\n- Corrective actions unrelated to the stated root cause.\n- Effectiveness checks consisting of a single clean result.\n- Spore formers or Gram-negatives attributed to the laboratory.\n- Isolates not identified, making source attribution impossible.\n- No concurrent environmental monitoring of the test session, so the laboratory hypothesis cannot be assessed either way.\n\nThat last one is worth emphasising. Without session monitoring data, a facility cannot support a laboratory-error conclusion even when it is true — the absence of evidence works against you in both directions.\n\n## After the Investigation\n\nCorrective action should address the identified cause, and effectiveness verification for a sterility positive is more demanding than for most deviations.\n\nExpect to run an additional media fill where an aseptic process cause is suspected, intensify environmental monitoring in the affected area for a defined period, and where a sterilisation cause is found, requalify the cycle rather than adjusting it. Closure should require a documented criterion agreed in advance, not a judgement made when the pressure to resume has built up.\n\n## How DSTechnoverse Helps\n\nWe support sterility positive investigations both as the testing laboratory and as an independent technical resource where a client's own laboratory generated the result.\n\nThat covers structured phase one and phase two protocols, organism identification and strain typing to establish or exclude a route, concurrent session environmental monitoring so a positive can be assessed against the conditions that produced it, sterility assurance review across sterilisation validation and media fills, impact assessment support, and investigation reports written to withstand inspection.\n\nWhere the evidence does not support invalidation, we say so plainly. That position protects a client far more than a convenient conclusion, and it is the one that holds up when an inspector reads the file two years later.\n\n## Frequently Asked Questions\n\n### Can we retest immediately?\n\nNot before phase one is documented. Suspicion of laboratory error is not a finding. Complete the investigation, identify the specific fault if one exists, and only then consider a retest.\n\n### How likely is a false positive?\n\nLess likely than most people assume. The test is far more prone to missing real contamination than to producing spurious growth, which is why the invalidation bar is high.\n\n### What if the organism is a common skin contaminant?\n\nIt is suggestive but not conclusive. Combined with a failed glove print or a documented breach, it supports a laboratory conclusion. With all controls valid, it does not — production personnel shed the same organisms.\n\n### Does a positive mean the whole batch is rejected?\n\nNormally yes, absent documented evidence of a test fault. The scope may extend further depending on what the batch shared with others in the campaign.\n\n### How long should the investigation take?\n\nPhase one within days. The full investigation commonly runs thirty days, extended with justification where typing or additional media fills are needed.\n\n### How should the investigation treat a positive from a small-batch product?\n\nThe compendial minimum for a small batch may be only four containers, which gives very little statistical power and makes a positive proportionally more alarming rather than less. With four units tested and one positive, the point estimate for the contamination rate is twenty-five per cent, and even allowing for wide confidence intervals the batch is in serious difficulty. Small-batch products also tend to be high value and clinically important, which raises the stakes on getting the disposition right. Resist the temptation to treat a small sample as weak evidence.\n\n### How much does a sterility positive actually cost?\n\nFar more than the batch. The direct loss is the product, but the consequential costs dominate: an investigation running thirty days or more, quarantine of any batch sharing the sterilisation load or campaign, at least one repeat media fill where an aseptic cause is suspected, intensified monitoring during effectiveness verification, and the management time of a multi-department team. Where product has already been distributed, notification and potential recall costs exceed everything else combined. This arithmetic is the argument for investing in the upstream controls — bioburden, environmental monitoring, media fill rigour — that make positives rare in the first place.\n\n### What if the same organism appears in two separate sterility positives months apart?\n\nType the two isolates against each other before anything else. If they are the same strain, you have a persistent source that has survived whatever corrective action followed the first event, and the investigation should start from the assumption that the original root cause was never found. If they differ, you have two independent events, which points to a systemic weakness in the aseptic process rather than a single niche. The two conclusions lead to completely different corrective actions, and only [strain typing](/blog/strain-typing-contamination-source-tracking) separates them.\n\n### Should we do a media fill after a sterility positive?\n\nWhere an aseptic process cause is suspected or cannot be excluded, yes. It is one of the few ways to demonstrate the process is capable, and it forms part of effectiveness verification.\n\n---\n\n**A passing sterility test is not proof of sterility — the assurance comes from the process.** DSTechnoverse provides [sterility testing](/services/sterility-testing) with method suitability validation, concurrent environmental monitoring and full positive-result investigation. Based in **Indore, Madhya Pradesh**, serving sterile manufacturers across India. [Talk to our microbiology team](/contact).","related":[{"slug":"media-fill-aseptic-process-simulation","title":"Media Fills and Aseptic Process Simulation: Designing a Test That Can Actually Fail"},{"slug":"sterility-testing-usp-71-guide","title":"Sterility Testing Under USP <71>: Methods, Validation and What the Test Can Prove"}]}