NIOSH researchers conducted a field survey at Veterinary Hospital G in August 2018. The purpose of the site visit was to identify and evaluate hazardous drug engineering controls as well as to sample for potential surface contamination at the hospital. NIOSH researchers also observed and interacted with the hospital's veterinarians and staff to obtain information about the hazardous drug work practices, daily activities, and oncology treatment processes. A TSI VelociCalc(TM) Plus Model 9565-P thermal anemometer was used to measure air velocities at the face of the biological safety cabinet (BSC), while a Wizard Stick handheld smoke generator was used to visualize air movement inside and around the periphery of the hood. Both the qualitative and quantitative tests showed that the BSC was operating appropriately. The BSC's average face velocity measured (0.51 m/s [100 fpm]) which is right at the minimum recommended face velocity of 0.51 m/s (100 fpm) for a Class II Type A2 BSC. The manometer function of the anemometer measured the oncology department's static room pressure to be negative. The pressure difference from the small kennel to the oncology room was negative with a pressure reading of -0.021 and -0.020 inches of water gauge (in. w.g.) with the oncology room's BSC blower on and off, respectively. The pressure difference from large kennel to the oncology room was negative with a pressure reading of -0.011 and -0.008 in. w.g. with BSC blower on and off, respectively. The pressure difference from the oncology office area to the oncology room was negative with a pressure reading of -0.027 and -0.024 in. w.g. with BSC blower on and off, respectively. These pressure measurements (excluding the -0.008 in. w.g. reading) meet United States Pharmacopeia (USP) <800>'s negative pressure requirements (-0.01 to -0.03 in. w.g.). A TSI Accubalance Plus Air Capture Hood Model 8373 was used to measure mechanically generated supply and exhaust airflows in the oncology room. The oncology room is designed to be under negative pressure and thus has more mechanical exhaust air (0.26 m3/s or 550 cfm) than supply air (0.22 m3/s or 472 cfm). The oncology room's volume (56.58 m3 [1998 ft3]) and mechanical exhaust airflow were used to calculate the room's ventilation rate as 16.5 air changes per hour (ACH) (with BSC exhaust blower deactivated). When the oncology room's BSC exhaust was activated, the overall room's exhaust rate increased by the amount of the BSC's exhaust, resulting a calculated ventilation rate of 26 ACH. Both of these ventilation rates exceed the minimum 12 ACH specified for unclassified containment secondary engineering controls. The presence of potential surface contamination was evaluated with wipe samples. These were collected in areas where the staff handled chemotherapy drugs within the oncology department. Wipe samples were also collected in less obvious places (i.e., telephone, door handles, floor of nearby restroom) to determine if current workplace safety practices at the hospital were adequate to prevent inadvertent contamination of these surfaces. In some cases, a single sample could be evaluated for more than one analyte simultaneously. The drugs administered during the NIOSH visit were vinblastine, vincristine, mitoxantrone, zoledronic acid injection, doxorubicin, Tanovea, asparaginase, and carboplatin. Sample results revealed that 4 of 6 wipe samples were positive for carboplatin (1.0 to 6.1 ng/sample). All of the 9 wipe samples were non-detectable (ND) for vinblastine. The ND determination means that contamination was either not present, or was present at levels below the detectable limit of the analytical method. Three of 14 wipe samples were positive for cyclophosphamide (1.6-3.9 ng/sample) while simultaneously being ND for doxorubicin, vincristine, methotrexate, and epirubicin. Two of the carboplatin and two of the cyclophosphamide wipe samples were between the LOD and LOQ. Nine out of 9 wipe samples submitted for N-methyldiethanolamine (MDEA) were positive (6.8 to 106 ng). Three out of 9 wipe samples submitted for toceranib, chlorambucil, and lomustine were positive for toceranib (2.2 to 1250 ng). MDEA was monitored as a potential stable marker for the highly unstable antineoplastic drug mustargen as explained in the text. Although many of the wipe sample analytical results were ND, there is no safe level of exposure when handling hazardous drugs. The presence of the carboplatin contamination is a reminder that the patients themselves can be a source of exposure, even when the drugs are not being directly handled. The cyclophosphamide (on surfaces), MDEA, toceranib, and methotrexate presence serves as two reminders: (1) that hazardous drug contamination can sometimes linger despite cleaning efforts and (2) the detected contamination on bathroom floor one might ordinarily think of as "safe", emphasizes the importance of proper work practices regarding the use of gloves and shoe covers, hand washing, and food/drink prohibitions within the hazardous drug handling environments. The detected contamination on the outside of the chemo transport bag serves as a reminder of the meticulous work practices required to avoid cross-contamination of surfaces expected to be "clean" as well as a reminder to treat all surfaces as potentially contaminated within the oncology treatment areas. Therefore, it is important to continue to use engineering controls (e.g., biological safety cabinets), supplementary controls (e.g., closed system drug-transfer devices), protective work practices (e.g., surface cleaning after every oncology patient, regardless of whether I.V. chemotherapy was administered), and personal protective equipment (e.g., gloves and gowns rated for chemotherapy protection, respirators, shoe covers, eye protection) to reduce unintentional exposures to the staff or pet owners.