Deployment of real-time particle detection monitoring system in operating theatres for airborne contamination assessments

Abstract

Aims

Surgical site infections (SSIs) are a major complication of orthopaedic implant surgeries, causing patient morbidity and reduced quality of life, and represent a substantial economic burden. Current methods for monitoring airborne contamination in operating theatres (OTs) are labour-intensive and delayed, limiting timely preventive actions. Advanced real-time monitoring technologies offer potential for improving infection control in surgical settings. This study evaluated real-time monitoring for airborne contamination; two scenarios were analyzed with the developed system: 1) the use of reusable non-disposable versus disposable surgical sheets; and 2) surgical team shift changes. SSI outcomes were also evaluated in relation to elevated particle levels.

Methods

This study was conducted in four OTs at Sahlgrenska University Hospital, Sweden. Particle counters were employed in each OT for detection of airborne contamination for continuous surveillance. SSIs leading to reoperations were extracted from national registries and integrated into the analysis.

Results

The use of reusable surgical sheets significantly reduced airborne particle concentrations across all sizes (0.5, 1, 5, and 10 µm; p = 0.022, p = 0.004, p = 0.009, and p = 0.015, respectively) compared with single-use sheets. Team shift changes were associated with increased airborne particle levels for 0.5, 1, and 5.0 µm (p = 0.001, p = 0.004, and p = 0.009, respectively). While smaller particle concentrations showed no consistent association with SSIs, larger particles (10 µm) were significantly elevated in SSI cases (p = 0.005 for maximum values and p = 0.009 for mean values).

Conclusion

Real-time monitoring systems proved effective in identifying factors influencing airborne contamination in OTs. Notably, non-disposable sheets outperformed disposable sheets in minimizing particulate dispersion, and surgical events with team shift changes showed an increase in maximum particle levels. Although the system shows promise for infection prevention and workflow optimization, its direct impact on SSI rates requires validation in larger cohorts. Future research should focus on integrating predictive algorithms and machine-learning to enhance clinical utility and drive improvements in surgical safety.

Cite this article: Bone Jt Open 2025;6(4):499–505.

Introduction

Surgical site infections (SSIs) are serious complications following orthopaedic implant surgeries.¹ These infections often lead to a diminished quality of life for patients, causing reduced mobility, psychological distress, and chronic pain.^2,3 SSIs impose a notable economic burden on healthcare systems due to prolonged hospital stays, additional treatments, readmissions, and need for more intensive care.^4,5 According to a 2009 report from the Centers for Disease Control and Prevention,⁶ the annual healthcare cost of SSIs in the USA was $3.5 billion. When broader societal costs are considered, such as lost productivity and long-term care needs, the total economic impact rises to approximately $10 billion annually.^6,7

The incidence of SSIs is expected to increase following orthopaedic surgeries and is driven by several inter-related factors. The global demand for orthopaedic procedures is increasing, due to an ageing population and the corresponding rise in the number of elderly patients requiring surgical intervention.^8-10 Orthopaedic implants provide surfaces conducive to bacterial biofilm formation, a mode of bacterial growth that shields pathogens from host immune defences and antimicrobial agents, complicating eradication efforts.¹¹ This challenge further increases the prevalence of antimicrobial-resistant pathogens, which undermines the effectiveness of prophylaxis and treatment strategies.¹² Together, these factors underscore the urgent need for innovative, proactive approaches to mitigate the burden of SSIs.^13,14

SSIs are multifactorial, with exogenous contamination widely recognized as a major contributor to wound infections during surgery.^15,16 Key sources of contamination include surgical personnel, door openings, and intraoperative traffic, all of which elevate airborne particle levels that may come into contact with the surgical site.^17,18 In the operating theatre (OT), conventional methods for assessing airborne bacterial contamination typically involve collecting and counting colony-forming units (CFUs) from air samples.¹⁹ However, CFU sampling methods are labour-intensive, disruptive to the surgical procedure, and typically take two to five days to obtain results, limiting the ability to implement timely interventions. Furthermore, these samples provide only a snapshot of microbial contamination in the OT and fail to capture temporal fluctuations.^19,20 A power analysis based on SSI rates of 1.5% to 2.0% indicates that a sample size of 10,000 patients is needed to assess an independent univariable, while approximately 70,000 patients are required for multivariable analysis, with a power greater than 80%.²¹ Hence, conventional CFU-based methods are not only inadequate for capturing the complexity of airborne contamination dynamics during surgical procedures, but are also unfeasible for large-scale studies, thereby limiting their utility in investigating the association with SSIs.

In contrast, industries such as pharmaceutical manufacturing, where airborne contamination also poses significant risks, have successfully implemented real-time monitoring systems to continuously track particulate levels.^22-25 These systems can provide immediate feedback, enabling swift actions when deviations from normal particulate levels occur. Given the similarities between clean room environments and OTs, where maintaining cleanliness is equally crucial, real-time monitoring technology holds considerable promise for healthcare applications.²⁶ The integration of such systems in other industries has demonstrated the value of instant feedback in detecting airborne contamination and prompting rapid interventions.²⁷ Translating these technologies to healthcare settings will require collaboration between engineers and clinicians to ensure the systems are appropriately adapted to meet the specific needs of the clinical environment.²⁸

This study introduces a monitoring system for airborne contamination, specifically designed for use during orthopaedic surgeries in OTs. The system integrates particle counters and environmental sensors, along with tracking of door openings and personnel presence through real-time video detection analysis. While similar configurations have been tested in simulated OT environments,²⁹ this system is uniquely engineered to capture and store real-time data throughout ongoing orthopaedic surgeries, building on successful application that has been introduced in other industries. The objective of this study is to identify scenarios associated with high particle emissions and assess their correlation with SSIs, providing clinicians with precise insights into contamination dynamics. By improving the understanding of these dynamics, the system aims to reduce the incidence of SSI and, ultimately, enhance patient safety and healthcare quality.

Methods

Location and study population

Data for this study were collected from four OTs at Sahlgrenska University Hospital in Gothenburg, Sweden. The study focused exclusively on surgical procedures involving the insertion of orthopaedic implants, as these procedures are particularly sensitive to airborne particle contamination due to the heightened risk of SSIs.

Materials and equipment

The OTs was equipped with ISO 17025-calibrated temperature and humidity sensors (ENV-THUM; InfraSensing, Belgium), differential air pressure sensors (ENV-AIRPRESSURE; InfraSensing) to measure pressure differences between the OT and adjacent corridors, and magnetic contacts (Standex-Meder, USA) to track door openings to the corridor. Additionally, OTs were equipped with ceiling-mounted fisheye cameras (M3067-P; AXIS, Sweden), with two cameras installed per room. To safeguard the privacy of patients and staff, a people-counting algorithm was applied to the camera data. This algorithm converted images into human detections with positional, size, and orientation data using the publicly available RAPiD neural network. Airborne particle concentrations were monitored using AeroTrak APC 6510 (0.5, 0.7, 1, and 5 μm) and AeroTrak APC 6301 (0.5, 1, 5, and 10 μm) (TSI, USA). The particle counters used comply to the ISO 21501-4 standard.

Sensor connectivity was established through a local network using Wi-Fi or ethernet cables, either directly or via sensor gateways. Collected data were stored in databases InfluxDB/MongoDB. The sensors used within the OTs are shown in Figure 1, which illustrates the system architecture, including two computers running containerized applications via Docker (USA) to facilitate development and deployment.

Fig. 1.

Overview of sensors (people-counting camera, particle counters, air pressure, temperature, humidity and door sensors) used within the operating theatre, with the system designed to seamlessly integrate additional sensors of interest. Data collected by these sensors is securely stored in a centralized database, enabling diverse applications and analyses. OP, operation; PC:s, computers.

Example assessment

This study investigated scenarios that may contribute to elevated particle counts within the OT: 1) the transition from disposable to reusable non-disposable surgical sheets; 2) team shift changes during ongoing surgical procedures; and 3) the association of particle levels with SSI. Scenarios one and two were selected as they represent common routine practices in OTs, with the potential to influence particle levels significantly.

In example one, maximum particle levels were measured to assess the immediate impact of multiple staff members entering the OT, anticipating a spike in particle counts at that time. In example two, mean particle levels were evaluated to capture the influence of the surgical sheets throughout the duration of the procedure.

Statistical analysis

The statistical analysis was conducted using JASP v.0.19.1.³⁰ Normality of the data distribution was assessed using the Shapiro-Wilk test, which evaluates whether the data deviate significantly from a normal distribution. Skewness and kurtosis were assessed via distribution and QQ plots as descriptive measures to complement the assessment of distributional characteristics. The assumption of homogeneity of variance was tested using the Brown-Forsythe test.

For independent univariate comparisons, a non-parametric Mann-Whitney U test was employed due to its suitability for ordinal data or non-normally distributed variables. Results are reported as medians and IQRs. Statistical significance was determined using a significance level (α) of 0.05.

Clinical follow-up and ethical approval

In this study, infections were defined as any surgical intervention performed to address an infection following the initial procedure. Data on reoperations due to infections were extracted from the Swedish Fracture Register and the Swedish Arthroplasty Register. These registries provide comprehensive, validated information on surgical outcomes, including postoperative complications such as infections requiring reoperation. All personal identifiers were anonymized and securely separated, ensuring they were never uploaded to the database.

The study was conducted following ethical principles for medical research and received approval from the Swedish Ethical Review Authority (2021-04805). The study protocol was registered with ClinicalTrials.gov (NCT05816135).

Results

Particle measurement for disposable versus non-disposable sheets

Particle concentrations (mean µm/m³) across four particle sizes (0.5, 1, 5, and 10 µm) were compared between disposable (n = 132) and non-disposable sheets (n = 132), as shown in Table I. For smaller particles, the introduction of non-disposable sheets was associated with a reduction in particle concentrations. Specifically, concentrations for 0.5 µm and 1 µm particles were notably lower with non-disposable sheets. A similar decrease in concentration was observed for larger particles, with a reduction in both 5 µm and 10 µm particle concentrations when non-disposable sheets was used. These differences were statistically significant across all particle sizes (p= 0.022, p = 0.004, p = 0.009, and p = 0.015, respectively).

Table I.

A comparison of airborne particle concentrations before and after the introduction of new surgical sheets in one operating theatre.

Particle concentration (mean µm/m³)	Disposable sheets (n = 132)	Non-disposable sheets (n = 132)	p-value*
0.5	5,327 (2,120 to 11,393)	3,248 (1,449 to 8,140)	0.022
1.0	2,194 (934 to 4,751)	1,146 (520 to 2,903)	0.004
5.0	334 (145 to 700)	221 (91 to 417)	0.009
10	127 (60 to 304)	86 (35 to 175)	0.015

Results are shown as median of mean values per operation (IQR).

Significance level at p < 0.05.

^*Mann-Whitney U test.

Particle measurements for team shift changes during ongoing surgeries

Comparison of particle concentrations (maximum µm/m³) across four particle sizes (0.5, 1, 5, and 10 µm) between operations involving shift changes (n = 468) and those without shift changes (n = 2,590) is presented in Table II. Operations with shift changes exhibited significantly higher median concentrations compared with those without shift changes. This trend was consistent across particle size 0.5 to 5 µm. For the largest particle size (10 µm), no significant difference was observed between the two groups (p = 0.430).

Table II.

Airborne particle concentrations during operations with and without shift changes for all operating theatres.

Particle concentration (maximum µm/m³)	Shift change (n = 468)	No shift change (n = 2,950)	p-value*
0.5	124,058 (43,849 to 429,898)	90,184 (32,509 to 345,618)	0.001
1	59,045 (18,372 to 167,406)	42,306 (15,256 to 130,035)	0.004
5	6,007 (2,525 to 14,127)	4,947 (2,297 to 11,177)	0.029
	Shift change (n = 199)	No shift change (n = 1,186)
10†	2,120 (795 to 6,007)	2,120 (1,060 to 4,594)	0.430

Results are shown as median of maximum values per operation (IQR).

Significance level at p < 0.05.

^*Mann-Whitney U test.

^†The particle counter (AeroTrak 6301) measuring 10µm was only installed in two operating theatres.

Particle levels association with SSI

Particle concentrations at sizes 0.5, 1, 5, and 10 µm were compared between patients who experienced no SSI (n = 2,992) and those who developed a SSI (n = 68), and can be seen in Table III.

Table III.

Association between particle concentration levels and surgical site infection for all operating theatres.

Variable	No SSI (n = 2,992)	SSI (n = 68)	p-value*
Particle concentration (maximum µm/m³)
0.5	93,300 (33,548 to 361,605)	116,854 (56,255 to 386,256)	0.066
1	43,463 (15,901 to 135,689)	52,512 (25,412 to 159,126)	0.100
5	5,156 (2,330 to 12,077)	6,288 (3,533 to 13,495)	0.170
Particle concentration (mean µm/m³)
0.5	7,844 (3,661 to 17,362)	7,595 (4,215 to 15,567)	0.794
1	4,101 (1,872 to 8,099)	3,782 (2,249 to 7,204)	0.745
5	531 (230 to 1,109)	529 (257 to 1,093)	0.773
Particle concentration (µm/m³)	No SSI (n = 1,363)	SSI (n = 22)	p-value*
Maximum value for 10†	2,120 (1,060 to 4,594)	3,534 (2,827 to 7,862)	0.005
Mean value for 10†	257 (90 to 625)	446 (282 to 927)	0.009

Results are shown as median of maximum and mean values per operation (IQR).

Significance level at p < 0.05.

^*Mann-Whitney U test.

^†The particle counter (AeroTrak 6301) measuring 10µm was only installed in two operating theatres.

SSI, surgical site infection.

For maximum particle concentrations, no significant differences were observed between the two groups for 0.5, 1, and 5 µm. However, a significant difference was noted for 10 µm, with higher concentrations in the SSI group compared with the non-SSI group (p = 0.005). Similarly, for mean particle concentrations, no significant differences were found for 0.5, 1, and 5 µm, however, the SSI group displayed significantly higher mean concentrations for 10 µm (p = 0.009).

Discussion

The objective of this study was to evaluate the implementation of a real-time monitoring system in OTs to generate data on air quality and its contributing environmental factors. The findings demonstrate that the system used in the study provides actionable insights into the factors that negatively impact air quality. Importantly, the real-time feedback enables proactive interventions that potentially can be used to reduce the incidence of SSIs. The results of this study indicate trends that larger particles (10 µm) are associated with SSI outcomes, as observed in both maximum (3,534 µm/m³ vs 2,120 µm/m³; p = 0.005) and mean (446 µm/m³ vs 257 µm/m³; p = 0.009) particle concentrations. It is also important to note that the Swedish Arthroplasty Register and the Swedish Fracture Register do not achieve full completeness for infection reoperations, and SSI outcomes may have been under-reported. Data collection for the current project is still ongoing, and we aim to include as many surgical events as needed to increase the estimated power for uni- and multivariate comparisons.²¹

The results from this study revealed that wash-and-reusable cloth sheets dispersed fewer airborne particles of all sizes (0.5 µm to 10 µm; p = 0.022, p = 0.004, p = 0.009, and p = 0.015, respectively) compared with disposable sheets. A systematic review focusing on orthopaedic and spinal surgeries reported no evidence of a difference between reusable and disposable sheets in reducing the risk of SSIs, emphasizing the need for further research.³¹ Our developed system in this study offers healthcare workers a practical tool for monitoring the impact of newly introduced medical equipment on particle load within the OT. Furthermore, it facilitates the efficient conduct of additional studies under clinically relevant conditions, supporting evidence-based decision-making regarding the suitability and effectiveness of new equipment before large-scale implementations.

While it may seem intuitive to associate team shift changes with increased particulate dispersion, subtle variations in intraoperative routines may go unnoticed without studies and supporting clinical evidence. This study found that particles ≤ 5.0 µm increased during team shift changes (p = 0.001, p = 0.004, and p = 0.009, respectively), which aligns with simulated studies.^32,33 A key advantage of the developed system is its capacity to provide real-time visualization of particle data directly to the OT staff. By offering continuous feedback on factors such as door openings and the presence of additional personnel, the system can enhance staff awareness of how their actions influence particle dispersion. This heightened self-awareness could promote behavioural adjustments and serve as a valuable educational resource, promoting practices that minimize particulate contamination.³⁴

Moreover, the system offers opportunities for retrospective analysis (e.g. by providing weekly reports to surgical staff). These reports can serve as a reflective tool, enabling teams to review air quality trends and identify underlying causes of any deviations. The system’s dual functionality (real-time intervention and retrospective assessment) makes it a versatile tool for enhancing OT practices and improving patient safety. However, the implementation of the current system in clinical settings presents challenges. The translation of clean room technology to healthcare requires adaptations to account for the complexity of OT environments, where human factors and different surgical protocols introduce variability.

By bridging the gap between traditional methods and emerging technologies, this system has the potential to reduce the economic and clinical burden of SSIs. The estimated cost of managing a single SSI with revision surgery ranges from $43,000 to $60,000,³⁵ and these costs are further amplified by indirect consequences, such as reduced patient quality of life and lost productivity. Consequently, SSIs contribute substantially to the overall strain on healthcare budgets worldwide. The estimated cost of implementing the proposed monitoring system, including the equipment used in this study, was approximately $11,000 to $12,000. The relatively low investment required for this system compared with the substantial cost of managing even a single infection underscores its potential economic benefit, and offers a cost-effective strategy for monitoring airborne contamination and helping to reduce the incidence of SSIs.

In conclusion, this study underscores the potential of real-time monitoring systems in OTs to assess airborne particle contamination following orthopaedic surgeries. The findings indicate that larger particles are associated with higher SSI rates. The system demonstrated its value by providing data that could prompt immediate interventions and promote behaviour awareness among surgical teams, potentially reducing SSI incidence. Furthermore, the system’s capacity for analysis offers opportunities for continuous improvement in surgical practices. The relatively low cost of the monitoring system compared with the high cost of managing SSIs suggests that this approach could offer significant economic and clinical benefits.

Take home message

- Real-time particle counting during orthopaedic surgery can help identify deviations in airborne particle levels, offering a potential tool to enhance the surgical environment and reduce the risk of surgical site infections (SSIs).

- However, further research is needed to determine particle level thresholds that correlate to an increased risk of SSIs.

Data Availability

The datasets generated and analyzed in the current study are not publicly available due to data protection regulations. Access to data is limited to the researchers who have obtained permission for data processing. Further inquiries can be made to the corresponding author.