Occupational Exposure Risk and Safety of Novel Intraperitoneal Oxaliplatin Delivery via Heated CO₂ and Dual-flow Nozzle System

Abstract

Background

Peritoneal carcinomatosis is challenging to treat, with poor prognosis. Although pressurized intraperitoneal aerosol chemotherapy (PIPAC) and hyperthermal PIPAC (HPIPAC) aim to improve patients' quality of life, there is a potential risk of antineoplastic drug exposure to healthcare workers during surgery. This study evaluates the risk of oxaliplatin exposure among healthcare workers during PIPAC and HPIPAC compared to systemic intravenous chemotherapy (IV).

Methods

Oxaliplatin was administered to a swine model via IV, PIPAC, and HPIPAC in the operating room. Surface samples (n = 484) were collected from the operating room and personal protective equipment (PPE), while air samples (n = 144) were taken near the respiratory tract. Platinum, as an indicator of oxaliplatin, was measured using inductively coupled plasma mass spectrometry (ICP-MS).

Results

Platinum was undetectable in 98.6% of the air samples, except one PIPAC leakage case. Surface contamination occurred in 8.7% of room samples and 13.9% of PPE, with gloves being most affected. Platinum contamination was also found on instruments like endoscope cameras and injectors, and cross-contamination on floors and injectors before PIPAC indicated suboptimal cleaning. No platinum was detected in the experimenter's blood and urine.

Conclusion

Airborne platinum exposure was minimal when no drug leakage occurred, but surface contamination, particularly on gloves and instruments, highlighted the need for stringent protective measures to reduce occupational exposure risk. Although the possibility of inhalation exposure was low and there are no legal exposure standards for oxaliplatin, precautions should be taken to avoid potential exposure.

1. Introduction

Peritoneal carcinomatosis is a common metastasis from other solid organ cancers, rarely originating from the peritoneum itself. Representing an advanced stage IV disease, peritoneal carcinomatosis has a poor prognosis, with survival rates of 15–45% at stages III and IV, compared to 70–90% at stages 1 and 2 [1]. Intravenous (IV) chemotherapy's limited efficacy is attributed to poor peritoneal vascularization and inefficient drug delivery across the plasma-peritoneal barrier, sparking interest in intraperitoneal administration (IP) which delivers the drug to the peritoneum and peritoneal cavity directly. IP is theoretically employed to achieve higher local drug concentrations in the peritoneal cavity, enhancing exposure to cancer cells. Hyperthermic intraperitoneal chemotherapy (HIPEC) was developed to increase drug exposure within the peritoneal cavity. This method uses a heated antineoplastic drug mixture circulated in the abdominal cavity.

However, challenges like limited drug distribution and penetration due to peritoneal nodules persist [2]. Pressurized Intraperitoneal Aerosol Chemotherapy (PIPAC) has been developed to enhance intraperitoneal chemotherapy delivery while minimizing systemic toxicity. However, concerns regarding occupational exposure due to aerosolized cytotoxic agents remain. A review of recent studies on PIPAC reveals consistent findings regarding airborne and surface contamination, as well as occupational health risks to medical personnel.

Airborne contamination was not detected in multiple studies. Larroque et al. [3] and Delhorme et al. [4] reported no measurable platinum in the air during oxaliplatin- and cisplatin-based PIPAC procedures, even in operating rooms lacking laminar airflow, when appropriate safety protocols were implemented. Similar findings were observed by Graversen et al. [5] and Solaß et al. [6], who reported undetectable airborne platinum and no increase in particle concentration. Reymond et al. [7] confirmed this through repeated environmental measurements across institutions. In an animal model study, Jung et al. [8] also found no airborne platinum during rotational PIPAC (RIPAC) procedures.

In contrast, surface contamination was frequently detected. Ametsbichler et al. [9] reported extensive platinum contamination on surfaces despite low airborne levels. Ndaw et al. [105] found platinum residues on surfaces around the application site. Jung et al. [8] detected platinum in 93% of surface samples collected from devices and personal protective equipment, although levels on masks, gloves, and shoes remained low.

Regarding occupational risk, most studies concluded that the risk is minimal under proper safety measures. Larroque et al. [3], Delhorme et al. [4], Graversen et al. [5], and Reymond et al. [7] found no evidence of environmental or biological exposure, and Solaß et al. [6] concluded that even personal respiratory protection was unnecessary. PIPAC has also been shown to be safe for patients when standardized protocols are followed [11]. However, some studies reported potential or actual low-level exposure. Ametsbichler et al. [9] emphasized the need for rigorous cleaning protocols, and Ndaw et al. [10] detected platinum in urine samples of healthcare workers, suggesting low but real biological uptake. Jung et al. [8] also warned of potential dermal exposure through surface contact, although respiratory risk appeared negligible.

The U.S. National Institute for Occupational Safety and Health (NIOSH) identifies inhalation, oral, and transdermal routes as potential exposure pathways for healthcare workers handling carcinogenic antineoplastic drugs. Although the International Agency for Research on Cancer (IARC) lists most Group 1 carcinogenic drugs as essential in cancer treatment, comprehensive exposure and risk assessments remain limited in the Republic of Korea [12,13].

Initial PIPAC safety protocols, developed in Germany under DEKRA, emphasized closed-abdomen procedures, remote control, laminar airflow, and PPE to prevent drug exposure. The International Society for the Study of Pleura and Peritoneum (ISSPP) later issued global guidelines via Delphi consensus, recommending key measures such as remote administration, ISO-class operating room ventilation, standardized personal protective equipment, checklists, and monitoring to ensure safe practice worldwide [7,14,15].

Hyperthermic pressurized intraperitoneal aerosol chemotherapy (HPIPAC), a novel approach that adapts the advantages of both HIPEC and the efficacious PIPAC, is under investigation [[16], [17], [18]]; however, occupational exposure assessments for healthcare professionals remain insufficient.

As HPIPAC is being investigated as a novel chemotherapeutic approach, it is essential to conduct further studies on occupational exposure to assess whether this method presents additional risks compared to PIPAC, and to evaluate whether the existing ISSPP safety protocols remain applicable or require further refinement [[16], [17], [18], [19], [20]].

This study aimed to assess occupational exposure to antineoplastic drugs in the air and surfaces during IV, PIPAC, and HPIPAC. By comparing exposure levels across these methods, the research seeks to evaluate the occupational safety of these innovative direct drug delivery techniques.

2. Materials and methods

2.1. Operation information

The operations were conducted using a swine model in a biosafety level 2 operating room at the Preclinical Experimental Center, Seoul National University Bundang Hospital (Bundang, Gyeonggi-do, the Republic of Korea). The procedure involved a swine model (triple inbred, microbial grade conventional, 30–35 kg body weight, male) treated with platinum-based oxaliplatin (5 mg/kg in 5% dextrose). The operations were repeated twice for IV and five times each for PIPAC and HPIPAC in a controlled operating room environment with an internal volume of 80.83 m³ (4.5 m × 6.0 m × 3.0 m) and conditions maintained at an average temperature of 21.0 ± 0.4°C and relative humidity of 57.2 ± 14.2%.

For the aerosolization of the oxaliplatin during PIPAC and HPIPAC experiments, the dual flow nozzle was used to make finer aerosol particles and a higher flow rate to enhance drug uptake and penetration into the peritoneum theoretically more than a commercially available hydraulic nozzle. For intraperitoneal hyperthermia in HPIPAC, the CO₂ used for the establishment of the pneumoperitoneum for operation was heated. We have developed a novel drug delivery system with enforced extracorporeal circulation of CO₂ to achieve these two goals [21].

Air currents were monitored using a smoke tester (No.500, GASTEC Co., Japan). Air velocity was assessed with a Velocicalc plus (TSI Inc., Shoreview, MN, USA), and lateral air supply measurements were taken using electronic balancing tools (EBT730, EBT731, TSI Inc., Shoreview, MN, USA) (Table S1, S2). The air flow dynamics within the operating room were characterized by an upward draft, with the average differential pressure measured at 12.6 ± 1.0 pascals. Air velocity was consistently low, ranging from 0.01 to 0.1 m/sec. Specifically, air was supplied from the left side at a rate of 0.2 ± 0.02 m³/sec, while it was exhausted on the right side at a rate of 0.06 ± 0.005 m³/sec. Air changes per hour (ACH) were calculated using the equation:

$$ACH\left(AirChangeperHour\right)=\frac{Q\left(\frac{m^3}{h}\right)}{V\left(m^3\right)}$$

The total air changes per hour (ACH) were calculated to be 23.0, which significantly exceeds the Ministry of Health and Welfare's standard of 15 for air purification facilities, indicating highly efficient air handling within the space.

For IV, diluted oxaliplatin was slowly infused into the antecubital vein over 2 min [22]. Unlike IV, where the antineoplastic drug is administered via intravenous infusion without any preparation, PIPAC is administered in four steps, and HPIPAC in five steps. First, five trocars are placed in the abdomen (two for CO₂ access, one for the nebulizer, one for the endoscopic camera, and one for the thermometer), and CO₂ is delivered into the abdominal cavity to inflate it. A nebulizer is connected to the trocar and positioned inside the abdominal cavity to spray the antineoplastic drug. One hundred mL of antineoplastic drug is injected at 20 cc/min and waited outside the operating room for diffusion in the abdominal cavity. After drug administration, the CO₂ in the abdominal cavity is filtered out through the installed evacuation system. For HPIPAC, after the first step, a thermometer is installed, and the CO₂ in the abdominal cavity is heated to approximately 42°C using an external heating device.

2.2. Sampling

2.2.1. Air sampling

Air samples were collected at three locations near the operating table, reflecting the 1.5 m standing height of healthcare workers, as shown in Fig. 1 [23]. A three-piece cassette system with a mixed cellulose esters membrane filter (MCE filter, 37 mm, 0.8 μm, SKC BGD) and a polycarbonate filter (PC filter, 37 mm, 0.4 μm, SKC Inc., PA, USA) was used. These were connected to an oil-free vacuum pump (WOB-L 2546, Welch Vacuum, Ilmenau, Germany) operating at 9 L/min, calibrated with a flowrate calibrator (Defender 510-H, Mesa Labs Inc., Lakewood, CO, USA) before and after each experiment. Background samples were taken for 30 min before the start of surgery (one per point), and the main samples were collected for 30 min after the start of surgery (three per point), covering the waiting period after drug spraying (Table S3).

Fig. 1.

Schematic diagram of the experimental operation room. Red circles with numbers indicate sampling locations.

2.2.2. Surface sampling

Surface sampling focused on areas with suspected antineoplastic drug exposure within the operating room, including medical devices and personal protective equipment (PPE), as indicated in Table S4 [24]. Room environment samples were collected from medical devices (nebulizers, electrocautery units, trocars, and endoscopic cameras), floors, trash can lids, and light handles. For PPE, surgical gloves, disposable plastic gowns, face shields, hair caps, and masks were included [25]. Each surface was sampled using Whatman 42 filter papers (ashless, grade 42, diameter 55 mm), pre-soaked in a wiping solution of 10% acetonitrile, 25% methanol, and 65% distilled water. An acrylic grid (10 cm × 10 cm) was used to ensure consistent surface coverage. Samples were transported to the laboratory in iceboxes and stored at low temperatures.

2.2.3. Bio sampling

The study participants were divided into two groups: the non-exposed group, consisting of two graduate students, and the exposed group, comprising the personnel who performed the procedures. Sampling for the non-exposed group was conducted concurrently with the bio-sampling of the exposed group.

Biological monitoring of urine and blood was performed before, immediately after, and on the day following the three administration types—IV (1 repetition), PIPAC (5 repetitions), and HPIPAC (5 repetitions). For PIPAC and HPIPAC, although each procedure was repeated five times, a total of 10 urine and 10 blood samples were collected at each time point because two staff members participated in each procedure.

For the exposed group, blood samples (4–5 cc) were collected twice: one hour before the experiment and immediately after its completion. The samples were stored in heparinized tubes, refrigerated, and then centrifuged and ultracentrifuged at 4 °C. The supernatant was stored in 5 mL snap-tubes at freezing temperatures. Urine samples (20 cc of midstream urine) were collected three times: one hour before the experiment, immediately after completion, and the following morning.

For the non-exposed group, both blood and urine samples were collected once, following the same collection and storage procedures described above.

No blood samples were collected on the day following the procedures for any administration type. All urine and blood measurements for both the exposed and non-exposed groups were below the limit of detection (<LOD). Therefore, they were not presented separately in the Results section.

2.3. Analysis

2.3.1. Analytical quantification

All samples, except those collected on PC filters, were analyzed for platinum using inductively coupled plasma-mass spectrometry (ICP-MS) following the NIOSH 7302 method. Platinum was chosen as a consistent indicator over oxaliplatin for methodological consistency. A platinum standard (10 μg/mL in 5% HCl, Agilent Technologies Inc., CA, USA) was diluted to 1, 2, 5, 10, and 20 ppb in 5% HNO₃, prepared from 70% nitric acid (ACS reagent, Sigma-Aldrich Co., MO, USA), to create a linear calibration curve (Table S5) [26]. PC filter samples were analyzed using FE-SEM (7800F Prime, JEOL Ltd., Tokyo, Japan) after carbon coating (3.21 × 10⁻² torr, 16 A) using a desk carbon coater (Turbo-pumped carbon coater, Vac Coat Ltd., London, England).

Surface samples, excluding those from air samples, masks, and hair caps, were digested in PTFE vessels with 5 mL of 70% HNO₃, then microwaved (MARS 6, CEM Corp., NC, USA). Masks and hair caps were cut into 25 cm² (5 cm × 5 cm) pieces using alcohol-disinfected scissors and pretreated similarly. After venting acid vapor, the contents were diluted with distilled water, further diluted 1:1, and filtered through a syringe with a PTFE filter. Pretreated samples were analyzed using ICP-MS (NexION 350D, Perkin Elmer Inc., TX, USA) (Table S6).

2.3.2. Quality assurance

The limit of detection (LOD) for platinum was calculated using the standard deviation of seven measurements at the lowest calibration standard of 1 μg/L, multiplied by three. The LOD was 7.24 ng/sample, with air samples ranging from 6.2 to 29.0 ng/m³ and surface sample ranging from 0.004 to 0.29 ng/cm². Surface sample results were expressed in μg/cm² with a correlation coefficient (r²) between 0.999 to 1.000. Recovery rates for platinum standards diluted to 5, 10, and 15 μg/L showed averages of 100.0 ± 3.9% for air samples and 98.6 ± 5.9% for surface samples across three trials, confirming robustness.

3. Results

3.1. Air sample

Table 1 summarizes the results of the analysis for the entire sample. In the air sampling analysis conducted using MCE filters during IV, PIPAC, and HPIPAC, platinum levels were below the limit of detection (LOD) in 142 out of 144 samples, representing 98.6% of all tests. Exceptions occurred during PIPAC, with platinum concentrations of 5.7 and 37.1 ng/m³, which were attributed to drug leakage while connecting the syringe to the injector (Table S7, Table S8). SEM and TEM analyses of air samples collected using PC filters during experiments involving only the aerosolized administration methods (PIPAC, HPIPAC) detected no platinum, even in cases of drug leakage (Fig. 2).

Table 1.

Detected number (detected sample/whole sample) and probability (%) of Pt in air and on surface samples due to antineoplastic drug exposure

Type	Air		Surface
			Room environment		PPE
	Background	Main	Background	Main	Background	Main
IV	0/6 (0.00%)	0/18 (0.00%)	—∗	—	0/21 (0.00%)†	1/36 (2.78%)
PIPAC	1/15 (6.67%)	1/45 (2.22%)	1/56 (1.79%)	7/43 (16.3%)	0/21 (0.00%)†	12/93 (12.9%)
HPIPAC	0/15 (0.00%)	0/45 (0.00%)	0/60 (0.00%)	10/54 (18.5%)	0/21 (0.00%)†	24/119 (20.2%)

^∗For IV administration, direct injection precludes room-environment exposure; thus, no samples were collected during background or injection.

^†For background of PPE, the same protective equipment was measured three times, independent of the antineoplastic drug application method, and the average value was used.

Fig. 2.

Scanning Electron Microscopy (SEM) and Energy Dispersive Spectroscopy (EDS) of polycarbonate filters used for air sampling during (a) PIPAC and (b) HPIPAC. While particulate matter is visible in the left and center images, EDS analysis (right) revealed no platinum peaks at the expected positions, indicating that these particles are not oxaliplatin or other platinum-containing substances.

3.2. Surface sample

Platinum was detected in 56 of 484 surface samples (11.6%) as shown in Table 3. Detection rates for IV, PIPAC, and HPIPAC were 1.9% (1 out of 54 samples), 8.7% (19 out of 218 samples), and 13.9% (37 out of 266 samples), respectively. PPE background samples were measured three times each, regardless of the operation type. The average values were calculated and included in the sample count for each operation type. Platinum contamination was primarily found on floors and injectors, with mean concentrations of 56.1 ± 39.1 ng/cm² and 66.4 ± 39.7 ng/cm², PIPAC, and 23.2 ± 11.1 ng/cm² and 16.9 ± 3.3 ng/cm² during PIPAC. For PPE, gloves had the highest contamination rates: 16.7% for IV, 28.6% for PIPAC, and 42.9% for HPIPAC (Table 2, Table S9, Table S10) (see Table 3) .

Table 3.

Geometric mean concentration (geometric standard deviation), range, and median in main surface samples with Pt detected (above the LOD, ng/cm²)

Type	Site	PIAPC				HPIAPC
		Number of samples (Detected/Total)	GM (GSD)	Range	Median	Number of samples (Detected/Total)	GM (GSD)	Range	Median
Room environment	Electrocautery	1/10	7.93	—	—	0/10	—	—	—
	Endoscope camera	1/14	1.06 × 103	—	—	2/20	25.1 (2.93)	11.8 – 53.7	—
	Floor	2/33	45.4 (2.34)	18.0 – 96.0	54.1	7/30	30.9 (1.67)	11.0 – 38.8	21.2
	Injector	2/16	60.2 (1.89)	38.3 – 94.5	—	3/20	16.7 (1.21)	13.9 – 20.5	16.5
	Light handle	1/6	30.7	—	—	0/10	—	—	—
PPE	Face shield	3/23	3.05 (24.6)	0.280 – 117	0.860	3/25	0.040 (1.42)	0.030 – 0.050	0.040
	Gloves	6/24	0.380 (9.53)	0.010 – 4.10	0.740	12/31	0.200 (13.0)	0.0100 – 42.9	0.400
	Gown	1/16	0.010	—	—	6/17	0.060 (2.23)	0.030 – 0.180	0.050
	Hair cap	1/23	0.890	—	—	0/24	—	—	—
	Mask	1/28	3.26	—	—	3/43	0.120 (3.10)	0.040 – 0.270	0.190

Table 2.

Number of detections (detected samples/total samples) and concentration (ng/cm²) of main surface samples by antineoplastic drug injection method (IV, PIPAC, and HPIPAC), expressed as mean ± standard deviation

Type	Site	IV		PIPAC		HPIPAC
		Number of detections	Concentration (ng/cm2)	Number of detections	Concentration (ng/cm2)	Number of detections	Concentration (ng/cm2)
Room environment	Electrocautery	—∗		1/10	7.93	0/10	—
	Endoscope camera			1/14	1.06 × 103	2/20	32.7 ± 29.7
	Floor			2/33	56.1 ± 39.1	7/30	23.2 ± 11.1
	Injector			2/16	66.4 ± 39.7	3/20	16.9 ± 3.32
	Light handle			1/6	30.7	0/10	—
	Trash can lid			0/9	—	0/10	—
	Trocar			0/14	—	0/16	—
PPE†	Face shield	0/22	—	3/23	39.2 ± 67.0	3/25	0.040 ± 0.010
	Gloves	1/9	0.260	6/24	1.31 ± 1.53	12/31	3.72 ± 11.8
	Gown	0/9	—	1/16	0.010	6/17	0.080 ± 0.060
	Hair cap	—	—	1/23	0.890	0/24	—
	Mask	0/14	—	1/28	3.26	3/43	0.160 ± 0.120

^∗Measurements were omitted for IV because the equipment was not utilized.

^†For the background of PPE, the same protective equipment was measured three times, independent of the antineoplastic drug application method, and the average value was used, which reflected several detections and concentrations of PPE.

4. Discussion

Airborne platinum, a surrogate for oxaliplatin exposure, was generally below detection limits, except for leakage during PIPAC. Safety protocols effectively minimized airborne contamination, though enhanced measures are needed for leakage scenarios. Existing protocols, adapted from the German guideline, lack contingency plans for such incidents, highlighting the need for stricter procedures and training.

Surface contamination rates of 8.7% (room environments) and 13.9% (PPE) suggest a need for caution despite the absence of legal oxaliplatin exposure limits. Gloves showed the highest contamination, aligning with previous studies (Table S11) [27,28]. Contaminated instruments (e.g., endoscope cameras, injectors) and floors indicate cross-contamination and inadequate cleaning.

Platinum (Pt) is commonly analyzed as a marker for anticancer drugs such as oxaliplatin or cisplatin, though other sources of exposure exist—most notably from automotive catalytic converters and dental materials. Ambient air Pt levels vary with traffic intensity. In Munich, levels rose from 7.3 ± 6.5 pg/m³ in 1993–1994 to 21.5 ± 13.8 pg/m³ in 1995–1996, peaking at 62 pg/m³ [29]. In Hong Kong, roadside air averaged 24.1 pg/m³, compared to 8.8 pg/m³ in suburban areas [30].

While platinum was detected at picogram levels in ambient air in the referenced study, it was not detected in the present study. This discrepancy may be attributed to two factors: (1) the operating room was a biosafety level 2 (BSL-2) environment, where air is supplied through high-efficiency filtration systems; and (2) the air sampling volume in this study was substantially lower than that in ambient air measurements, resulting in an elevated limit of detection (LOD). Accumulation in road dust and soil has also been reported, with Seoul showing 85–310 ng/g in road dust and 30–160 ng/g in roadside soil, further supporting the role of automotive catalysts as a primary source [31].

However, biomonitoring data suggest that environmental Pt levels may not be the main determinant of internal exposure. In the German Environmental Survey (GerES III), urinary Pt concentrations were most strongly associated with the number of noble metal dental alloy restorations (NMDARs), with each additional tooth linked to a 35.6% increase. Other factors—such as chewing gum, coffee consumption, and socioeconomic status—also influenced Pt levels, while traffic exposure had no significant effect [32]. Similarly, a study in Rome found no significant association between urinary Pt levels in children and traffic exposure [33].

While no ambient air quality standard for Pt exists, occupational exposure limits have been established: 1 mg/m³ for metallic Pt (ACGIH, NIOSH) and 0.002 mg/m³ for soluble salts (ACGIH, OSHA, NIOSH). Germany's DFG MAK committee limits the peak concentration of platinum in chloroplatinates to 0.002 mg/m³, and ACGIH has proposed a Biological Exposure Index (BEI) for urinary Pt at 0.01 μg/L [34].

Pt accumulation in urban environments is evident, reflecting widespread emission from catalytic converters, industrial combustion, and medical waste. Studies consistently show elevated Pt levels in road dust, surface soil, and urban aerosols, particularly in high-traffic areas, indicating continuous environmental input [[30], [31], [32], [33]]. However, despite the presence of Pt in the environment, its contribution to internal human exposure appears limited. Biomonitoring evidence suggests that personal and medical factors—such as dental alloys, chemotherapy exposure, or lifestyle habits—play a more significant role in determining urinary Pt concentrations. This highlights that environmental Pt serves more as a background indicator rather than a direct health determinant. Nevertheless, ongoing environmental and biological monitoring is warranted to assess potential cumulative exposure and emerging risks associated with increasing Pt use in medical and industrial applications.

Biospecimens were collected once for each injection method (IV, PIPAC, and HPIPAC) by two medical and two non-exposure personnel. The results suggest that oxaliplatin does not enter the body through inhalation or surface exposure, as it was not detected in any blood or urine samples. However, as the study indicates, oxaliplatin was found on some medical equipment and PPE surfaces, but the use of personal protective equipment effectively prevents it from entering the body.

Previous studies have reported varying limits of detection (LOD) or quantification (LOQ) for platinum, depending on the matrix and analytical approach. In 2013, Solaß et al. reported an LOD of 0.3 μg/sample for platinum in cisplatin-based PIPAC [6]. Graversen et al. (2016) presented an airborne LOD of 100 ng/filter, without specifying the air volume [5]. Ndaw et al. (2018) reported LOQs of 100 pg/wipe sample and 10 ng/L in urine [10]. Ametsbichler et al. (2018) reported LODs of 0.02 ng/wipe sample and 6 pg/air sample (≈3.1 pg/m³) [9]. Larroque et al. (2021) reported LODs of 5 ng/L in plasma and 3 ng/L in urine [3]. Jung et al. (2025) reported an LOD of 0.37 ng/sample for air samples, though airborne concentration was not calculated [8]. In the present study, the LOD was 7.24 ng/sample, corresponding to 6.2–29.0 ng/m³ depending on the sampled air volume. These results reflect substantial variability in detection sensitivity across studies and emphasize the need to optimize both analytical techniques and sampling strategies (e.g., increasing flow rate or sampling time) to improve detectability of airborne platinum during PIPAC procedures.

This study's limitations include the partial identification of contamination sources, such as surgical instruments and indoor environments, without fully accounting for all potential contributors to overall exposure levels. To address this, efforts were made to identify broader contamination sources, including cross-contamination and inadequate cleaning practices. Additionally, the study lacked sufficient longitudinal data to comprehensively assess the chronic exposure risk. While significant airborne and surface contamination was not observed, the possibility of repeated exposure to hazardous substances cannot be ruled out. Therefore, future research should extend monitoring periods to provide a clearer understanding of occupational exposure risk. better. Variations in surgical environments and practices across different hospitals may have introduced inconsistencies in exposure levels, limiting the study's generalizability. However, these variations were addressed by the research team that pioneered the PIPAC technique, thereby minimizing inter-study variance.

In summary, this study evaluated occupational exposure to antineoplastic drugs during IV, PIPAC, and HPIPAC using a swine model.

Airborne platinum—an indicator of oxaliplatin exposure—was below the LOD in 98.6% of samples. The only exceptions occurred during one incident of oxaliplatin leakage during PIPAC, where platinum concentrations were 5.7 and 37.1 ng/m³. These incidents highlight the need for stringent containment measures and continuous monitoring to mitigate inhalation exposure risk under abnormal conditions. The highest concentrations were detected during PIPAC, underscoring the necessity of stringent cleaning protocols and enhanced management to mitigate surface risk.

In conclusion, although most air and surface samples demonstrated minimal exposure to oxaliplatin, this study underscores the need for enhanced safety protocols, particularly in managing drug leakage and surface decontamination. Continued compliance with stringent safety measures, coupled with the advancement of real-time monitoring technologies, is essential to protect the health and safety of healthcare workers. Ultimately, this study emphasizes the need to protect healthcare workers from harmful substances, while simultaneously advancing cancer treatment.

CRediT authorship contribution statement

Soyeon Lee: Writing – original draft, Visualization, Methodology, Conceptualization. Jongmin Park: Validation, Data curation. Myoungho Lee: Resources, Formal analysis. Mira Yoo: Validation, Project administration. Chanmi Bang: Formal analysis, Data curation. Sungho Hwang: Project administration, Investigation. Dukyeon Kim: Investigation, Data curation. Soo Kyung Bae: Formal analysis, Data curation. Hyungho Kim: Project administration, Funding acquisition, Conceptualization. Chungsik Yoon: Writing – review & editing, Supervision, Investigation.

Animal research (ethics)

The preclinical experiment using the porcine model was approved by the Institutional Animal Care and Utilization Committee of Seoul National University Bundang Hospital (IACUC No. BA-2304-366-010-01). All experiments were performed in accordance with relevant guidelines and regulations.

Statement on the Use of AI Tools

During the preparation of this manuscript, the author(s) used ChatGPT in order to improve readability and clarity. After using the tool, the author(s) reviewed and edited all content as necessary and take(s) full responsibility for the final content of the publication.

Conflicts of interest

The authors declare no conflict interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article.