What 'lack of assurance of sterility' means in a drug recall

Since June 2012, 6,094 drug recall records in FDA's enforcement data have given a lack of sterility assurance as their reason, about one in three of the 18,076 records published through Oct. 28, 2026, according to GxRadar's analysis of openFDA¹. FDA put 6,082 of them in Class II, its middle tier.

A much smaller group of records uses a different word. Only 153 cite "non-sterility," meaning contamination was actually found, and FDA classified 148 of those as Class I, the tier for products with a reasonable probability of causing serious harm or death.

The gap between those two phrases explains a great deal about how sterile medicines are regulated. A lab test cannot prove a batch is sterile, so FDA judges sterility by whether the manufacturing process can be shown to deliver it. When a plant cannot show that, the batch can be recalled even if no microbe has turned up. This piece explains the testing limits behind that rule, the 2004 FDA guidance that governs aseptic manufacturing, and what 14 years of recall data show.

Why a passing sterility test proves little

CGMP rules require lab testing of every batch of a drug "purporting to be sterile," under 21 CFR 211.167(a)⁶. The test samples only a small number of units. FDA's 2004 aseptic processing guidance³ quotes the U.S. Pharmacopeia's statistical warning that the sampling plan "only enables the detection of contamination in a lot in which 10% of the units are contaminated about nine times out of ten."

FDA then gives its own illustration. If a 10,000-unit lot with a 0.1 percent contamination rate were tested using 20 units, "there is a 98 percent chance that the batch would pass the test." For that reason FDA treats any positive result as "a serious CGMP issue," and says an initial positive can be set aside only when the growth "can be unequivocally ascribed to laboratory error."

Assurance has to come from somewhere else. Section 211.113(b)⁵ requires written procedures to prevent microbiological contamination of products purporting to be sterile, and says those procedures "shall include validation of all aseptic and sterilization processes." Section 211.42(c)(10)⁷ requires aseptic processing areas to have HEPA-filtered air under positive pressure, a system for monitoring environmental conditions, and a system for cleaning and disinfecting rooms and equipment. A recall for lack of sterility assurance often means one of those links in the chain could not be documented.

Aseptic processing and terminal sterilization

There are two ways to make a sterile product. In terminal sterilization, the filled and sealed container goes through heat or irradiation at the end. In aseptic processing, the drug, container and closure are sterilized separately and then brought together in a highly controlled room, with no final kill step. The guidance says aseptic processing "involves more variables" and calls it "a well-accepted principle" that sterile drugs should be made aseptically "only when terminal sterilization is not feasible." A footnote adds that nearly all drugs recalled for nonsterility or lack of sterility assurance from 1980 to 2000 were aseptically processed.

Where sterile product is exposed, the critical zone must meet Class 100, or ISO 5, air cleanliness. Plants prove their aseptic lines with media fills, running sterile growth medium through the process in place of the drug and incubating every unit to see whether anything grows. FDA recommends at least three consecutive successful runs to qualify a new line, then semiannual requalification, and calls 5,000 to 10,000 units "a generally acceptable starting point" for run size.

Source: FDA, Sterile Drug Products Produced by Aseptic Processing (2004), section IX. The guidance also says that for any run size, intermittent contamination "can be indicative of a persistent low-level contamination problem." Its guidance page⁴ still lists the 2004 document as current.

What the recall records show

Counting records whose reason text includes "sterility assurance" or "assurance of sterility," the share swings widely from year to year, because a single facility recalling its whole product line can add hundreds of records at once.

Source: GxRadar analysis of openFDA drug enforcement records by report date. Each record is one product; a single recall event can contain many.

Injectables dominate the product mix. Of the 6,094 records, 3,813, or 63 percent, describe the product with the words injection, injectable, syringe, vial or infusion, and 422, or 7 percent, mention ophthalmic or eye products. Many of the largest bursts trace to compounding operations, which our explainer on 503A and 503B compounders covers, and to eye drops, as in the 2023 recalls.

Reason text is often terse. In a sample of 100 of the 153 records from 2026, 54 said only "Lack of Assurance of Sterility." The rest pointed to FDA inspection findings at the plant or a contract manufacturer, processing equipment used for compounded products, and packaging defects such as wrinkled lidding, open seals or ports that could leak.

FDA's own quality office reports similar patterns. Its FY2024 Report on the State of Pharmaceutical Quality¹⁰ found that sterility assurance issues made up 28 percent of contamination-related recalls that year. The FY2025 report¹¹ said contamination and sterility problems accounted for 13 percent of the 4,059 initial Field Alert Reports manufacturers filed, and that such defects fell 20 percent from the year before.

Why most of these recalls are Class II

Class definitions in 21 CFR 7.3² help explain the split. Class II covers situations where use of a product "may cause temporary or medically reversible adverse health consequences" or where "the probability of serious adverse health consequences is remote." A lot with documentation gaps but no organism found has often been placed there. When contamination is confirmed, the records show FDA usually moving to Class I.

Confirmed cases show why the documentation matters. In 2012, contaminated steroid injections made by a compounding pharmacy caused a multistate fungal meningitis outbreak that CDC counted¹² at 753 cases and 64 deaths in 20 states. Our account of that outbreak and the law that followed traces the regulatory response. FDA's 2019 public warning guidance¹⁴ lists sterile injectables that may contain particulate matter among its examples of recalls that may warrant a public warning.

Inspections and supply

Sterility controls are also among the most common inspection findings. In fiscal 2025, FDA investigators cited section 211.113(b) 105 times on drug-site Form 483s, 68 for missing or unfollowed procedures and 37 for inadequate validation, and cited the aseptic area requirements in section 211.42(c)(10) 67 times, according to FDA's fiscal 2025 inspection observation file⁹ published on its inspection observations page⁸. The FY2025 quality report named inadequate control of microbiological contamination as one of three top citations behind facility-related holds on drug applications.

When sterile lines stop, supply can stop with them. As of Nov. 3, 2026, 71 percent of the 69 current shortages on FDA's list involved injectable products, according to GxRadar's analysis of openFDA shortage data¹³, and our piece on sterile injectable shortages explains why. Anyone unsure whether a recalled lot affects a medicine they use can ask a pharmacist, and the GxRadar recall tracker lists sterility and contamination recalls as they are classified.