FDA Form 483

Regulatory Event

Between March 30 and May 12, 2026, FDA district offices in Detroit, Nashville, Dallas, and the Los Angeles District’s Owings Mills and Irvine field offices closed inspections at five independently operated compounding and outsourcing facilities: Pharmaneek Inc (Indianapolis, Indiana), Wells Pharmacy, Inc. (Dyersburg, Tennessee), Richmond VA Medical Center (Richmond, Virginia), VA San Diego Healthcare System (San Diego, California), and Central Alabama Veterans Health Care System – West Campus (Montgomery, Alabama). Collectively, investigators documented 32 observations across the five Forms 483 — 3, 2, 8, 11, and 8 respectively. None of these five sites currently carries an identified FDA Warning Letter; these are inspectional observations, not enforcement actions, and each facility retains 15 business days from its own inspection close-out to respond in writing. What makes this group worth analyzing together is not any single site’s risk profile, but the pattern that emerges when five unrelated facilities, inspected within a six-week window, are cited for strikingly similar categories of gaps.

The clearest pattern is smoke study rigor. Three of the five sites — Richmond VA, VA San Diego, and Central Alabama VA — received observations stating that dynamic smoke studies for ISO 5 laminar airflow hoods, biological safety cabinets, and laminar airflow workbenches did not adequately represent actual production conditions. VA San Diego’s finding was the most extensive: smoke study videos ranging from just 16 to 58 seconds, across four hoods and two segregated compounding areas, that failed to introduce smoke around the actual equipment and materials staged during real compounding — scales, cameras, sterile prep mats, syringes, closed transfer devices, and vial reconstitution mixers. Central Alabama VA’s smoke study used only two vials and one syringe to simulate a process that, in routine compounding, involved a bin, drug vials, a saline bag, a syringe, a needle, gloves, a sterile fluid bottle, and wipes — and jetted smoke from under a foot away rather than dispersing it to visualize turbulence or dead space. Richmond VA’s smoke studies similarly used an inadequate smoke volume and did not represent dynamic manipulation. FDA’s 2004 Aseptic Processing Guidance treats the dynamic smoke study as the qualification step that demonstrates unidirectional first air is preserved once real components and human intervention enter the ISO 5 critical zone — a static or under-loaded smoke study cannot support that conclusion, regardless of how compliant the hood’s certification paperwork looks.

What the Record Documents

The second pattern is first-air and gowning discipline observed directly during live compounding — not inferred from records, but watched by an investigator in real time. At Richmond VA, a technician’s hand was observed blocking the airpath between the High-Efficiency Particulate Air (HEPA) filter face and the critical work surface while injecting components into a total parenteral nutrition (TPN) bag, and another technician leaned into the ISO 5 hood with exposed forehead and eye-area skin. At VA San Diego, operators rested elbows on the ISO 5 work surface and biological safety cabinet front grille, moved quickly enough to disrupt airflow while spraying and drying gloved hands mid-process, and wore sterile gloves manufactured in January 2022 during active compounding. At Central Alabama VA, technicians wiped materials with non-sterile disinfectant in an unclassified segregated compounding area and moved them into the ISO 5 zone without re-sanitizing. Investigators watching aseptic technique breaches occur live, across three unrelated facilities, in the same six-week window, signals that this is where FDA’s current compounding-inspection emphasis sits — not paperwork review, but direct behavioral observation of the technicians actually performing the manipulations.

The third pattern is hazardous-drug and beta-lactam cross-contamination control. At Pharmaneek, a hazardous drug (Hydroxyurea oral suspension) was cleaned from reusable utensils using a liquid cleaner with no deactivating agent, and the washed utensils were then returned to a shared bin also used for non-hazardous production. At VA San Diego, multiple beta-lactam drugs — ceftaroline fosamil, ampicillin, ampicillin/sulbactam, ertapenem, and two strengths of meropenem — were produced in the same ISO 5 hoods as non-beta-lactam products, with cleaning frequency and inactivating-agent contact time that investigators judged inadequate to prevent cross-contamination. Beta-lactams carry a well-documented cross-sensitization risk, and USP General Chapter <797>, the United States Pharmacopeia’s sterile compounding standard, together with FDA’s compounding guidance, expects either physical segregation or a validated cleaning and inactivation process specific to that risk — a shared hood with generic cleaning is not equivalent to either control.

Technical and Quality Context

A fourth, narrower pattern touches non-sterile hazardous and hormone-pellet process validation. Wells Pharmacy’s 483 documents Estradiol and Testosterone pellet lots released with 28% and 43% out-of-specification pellet weights, with no investigation performed before release, alongside a cleaning validation study that tested only two of the facility’s hazardous drug products and used a different number of cleaning passes than the currently used procedure. Process and cleaning validation exist to establish, in advance, that a manufacturing process reliably produces conforming product; reviewing out-of-specification rates after the fact, without investigating root cause, does not substitute for that upstream validation work.

For quality leaders at any sterile or hazardous-drug compounding site, the actionable takeaway from this trend is specific. Dynamic smoke studies should be re-run whenever component types, fill volumes, or intervention patterns change — not treated as a one-time qualification — and should be observed, ideally by an independent reviewer, under conditions that mirror actual daily compounding load. Gowning and first-air discipline should be assessed through live, unannounced observation of working technicians, not solely through record review, since that is now demonstrably how FDA is inspecting these facilities. And any site producing both hazardous and non-hazardous, or beta-lactam and non-beta-lactam, products in shared equipment should independently verify that its cleaning validation covers every product actually run through that equipment, at the cleaning procedure currently in use, not the procedure in place when the validation study was originally performed.

Decision Relevance

XGene Consulting helps hospital pharmacies, 503A and 503B compounding facilities, and sterile manufacturers redesign smoke study protocols, conduct live aseptic-technique assessments ahead of FDA inspection, and close cleaning and process validation gaps for hazardous and beta-lactam product lines. If your site’s last smoke study predates a change in your product mix or fill equipment, that gap is worth closing now.