Improving ventilation hygiene in wards for immunocompromised patients: a workflow comparison of traditional and innovative duct cleaning methods

Summary

Introduction

For a vulnerable hospital population, hospital ventilation systems can pose a risk if not maintained correctly. Dust accumulation in air ducts can create environments that allow fungal spores, such as Aspergillus.spp, to thrive, potentially leading to severe infections like invasive aspergillosis. Our study aimed to develop a safe protocol for cleaning ventilation systems on an active ward in healthcare settings.

Methods

We conducted a comparative evaluation of ventilation duct cleaning in adjacent hospital rooms. In one room, the ducts were accessed from within the room using a traditional brushing method, while in the other room, the ducts were primarily accessed from outside using a novel method including the use of newly developed low speed brushes and point extraction of debris. Air quality was monitored by counting particles every 15 minutes in 100-liter samples using a calibrated air sampler and collecting microbiological specimens before and after the cleaning process.

Results

In-room cleaning caused a significant spike in airborne particles of all sizes, along with a modest increase in Aspergillus spp. spores, which took an hour to return to baseline levels. Conversely, external access for cleaning did not notably impact room air quality.

Conclusion

These findings highlight the risks associated with internal duct cleaning, especially for patients who are vulnerable to airborne fungal infections. External duct access proves to be a safer alternative, ensuring minimal disruption to the air quality in patient care areas. This study supports the necessity of strategic planning in hospital ventilation maintenance to protect vulnerable populations.

Introduction

Ventilation systems are a critical component of the hospital infrastructure, ensuring the circulation of clean air in patient care areas and reducing the risk of airborne transmission of infectious agents. However, over time, these systems can become heavily contaminated with dust, mold, and other particulate matter. Accumulation of such contaminants within ventilation ducts can pose significant health risks in immunocompromised patients [1]. One of the most concerning pathogens in this context is Aspergillus fumigatus, a type of fungus that can thrive in dust and debris, potentially leading to severe invasive infections in vulnerable individuals. Perdelli et al. [2] demonstrated that a well-maintained ventilation system can significantly reduce the environmental load of Aspergillus spp. spores.

Haematology patients are especially at risk due to their weakened immune systems. Outbreaks of Aspergillus infections in hematology wards and other wards with immunocompromised patients have been documented, highlighting the importance of maintaining clean and safe ventilation systems [1]. However, current practices vary and evidence-based guidelines to minimise airborne contamination risks are lacking, underscoring the need for standardised protocols to ensure safe duct maintenance.

Routine duct cleaning is an established practice to mitigate these risks [3]. Mechanical cleaning is the most commonly used method to remove accumulated contaminants and is typically performed with brushes, rotating brush heads, or air compressors to detach dust, debris, and biofilm from the duct walls [4]. Technicians typically access the ventilation ducts through existing openings, such as ventilation covers within patient rooms.However, these traditional cleaning techniques may not always be suitable or effective in hospital environments with stringent infection prevention requirements. Furthermore, completely evacuating patient areas to facilitate cleaning is often unrealistic due to the constant care needs of hospitalised patients, particularly those in critical or haematology units, and the high admission pressure limiting the availability of empty rooms.

Ghent University Hospital, a large teaching hospital in the Flanders region of Belgium, comprises multiple buildings, with diverse ventilation systems, or, in some areas, none at all. The haematology department, housed in a building dating from the 1970s, is equipped with a ventilation system and consists of 10 single and a double patient room(s) and eight stem cell transplant (SCT) rooms. The SCT rooms are equipped with horizontal laminar airflow through HEPA filters and provide 33 air changes per hour. The SCT rooms are maintained under positive pressure. This study, however, focuses on the standard rooms having a conventional ventilation system without HEPA filters, providing only three to four air changes per hour.

Visual inspection revealed significant duct contamination necessitating cleaning. Ideally, patient areas would be cleared during this process, but this was impractical, as the rooms are specifically designed for the care of this vulnerable population. Transferring patients to another location was not a feasible option, nor was delaying duct cleaning, given the increased risk of fungal contamination and potential Aspergillus spp. spread.

Due to the inability to relocate haematology patients and the urgent need for intervention, an alternative, safer cleaning method had to be identified. Traditionally, ducts are accessed for cleaning through existing openings, often via the ventilation grilles in patient rooms. However, this approach does not eliminate the risk of dust dispersion, posing a potential hazard to patients. Our study aimed to investigate whether effective duct cleaning could be achieved by accessing the ducts via the main duct outside the patient rooms, and, more importantly, whether this approach would have less impact on air quality and therefore be safer to patients. To assess this, a comparative analysis of both methods was conducted.

Methods

To evaluate the impact of different ventilation duct cleaning methods on air quality, a test was conducted in a department with identical infrastructure to the haematology unit but without immunocompromised patients. Two adjacent, unoccupied rooms were selected, where duct contamination levels were comparable to those in the haematology department, as visual inspection confirmed. Both rooms were located on different floors of the same building where the ventilation ducts are connected to the same ventilation system. Ventilation duct cleaning was performed in accordance with EN 15780 standards [5].

In one room (Figure 1, room A), the ducts were accessed internally (method A), using standard cleaning brushes at normal speed. While in the adjacent room (Figure 1, room B), the ducts were primarily accessed externally, from outside the room using a novel method where lower brush speeds were applied and point extraction, integrated into the brush system, was provided (method B). For method B, access points were created in the main duct located in the ward corridor. The hypothesis was that method B would be the safer option, minimizing dust dispersion in the patient environment.

Figure 1.

Technical floor plan of room A & B with ventilation extraction ducts.

Cleaning procedure

In room A, standard cleaning brushes were used with conventional brush speeds. In room B, the cleaning service provider (Hamster Cleaning, Belgium) employed the 95G method, a complex technique differing significantly from standard duct cleaning methods. Neither cleaning methods required deactivation of the ventilation system. When cleaning extraction ducts, it is preferable to extract any dislodged debris as quickly as possible.

The 95G method is developed by Hamster Cleaning and is based on 3 pillars, an assessment phase to determine the exact approach, a cleaning phase and a management phase documenting the past and future cleaning actions. During the cleaning phase, the technician uses specialised brushes with ultra-low linear and rotational speeds, specifically designed for controlled debris removal (Hamster Cleaning, patent pending). Instead of applying general negative pressure, point extraction was used (Brushbeast, Rotobrush), meaning that released dust was immediately extracted at the brush location of the brush itself, reducing the risk of airborne dispersal. (Figure 2).

Figure 2.

Ultra-low speed brush with point extraction in action. The blue arrows indicate the direction of airflow. The brush operates at an ultra-low linear and rotational speed while simultaneously extracting released dust at the point of contact, minimizing airborne dispersion and ensuring controlled debris removal. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article).

During method A (internal access), the technician entered the room with the required materials, dusted the ventilation grille, and then opened it to access the ventilation duct (extraction). The cleaning brush was inserted through the opening, and the duct was cleaned with standard brush speeds. In method B (external access) the 95G brushing method was used with lower brush speeds. The technician dusted the ventilation grille inside the room but did not open it. Instead, the cleaning brush was introduced through a newly created access point in the main duct outside the room, positioned above the ventilation grille in the room.

After air duct cleaning, the rooms were sanitised by the regular housekeeping staff to remove any potential dust dispersion.

Air quality and microbiological assessments

To compare both methods, particle measurements were performed before, during, and after the cleaning process. Particles were measured at regular intervals of 15 minutes using a calibrated particle counter (Aerotrack 9500, TSI), simultaneously in both rooms at the patient's bed position. This resulted in 18 samples measured per room (36 samples in total). Additionally, microbiological air sampling was conducted before and during the cleaning procedures using a calibrated air sampler (P100 Microbial Air Sampler, EMTEC), also at the patient's bed position. Each of the four samples consisted of 1000 liters of air drawn through Tryptic Soy agar with 5% sheep blood (BD, Erembodegem, Belgium), which was then incubated for at least 48 hours at 37°C.

Bacterial colonies were counted manually after 48 hours. Aspergillus spp. colonies were counted after 48 hours and again after 5 days. The number of CFUs per liter of air sampled was inferred from the colony count. Identification of the species was performed with MALDI-TOF MS (Bruker, Germany).

Data recording and analysis

The results were recorded in a comparative table in MS Excel for analysis. No statistical analysis was performed, as the study was conducted for practical evaluation purposes rather than hypothesis testing.

Results

Figure 3 shows the particle count over time, measured before, during, and after ventilation duct cleaning in Room A (red) and Room B (green). Particle size measurements focused on 3 μm particles, as this size is comparable to Aspergillus fumigatus spores.

Figure 3.

3μ particle count results over time.

These measurements were taken simultaneously in both rooms, allowing for the assessment of any potential cross-room influence of the cleaning activities on air quality.

The red line represents the results for Method A (in-room cleaning), which led to a significant increase in airborne particle concentrations. The particle count peaked during cleaning and required approximately one hour to return to baseline levels. Importantly, there was no measurable impact on air quality in the adjacent room (Room B), indicating that contamination remained localized. In contrast, the external cleaning method (Method B, green line) did not lead to a significant change in airborne particle concentrations in either room, suggesting that this method effectively prevented airborne dispersion.

Following the air duct cleaning, particle counts increased again in both rooms during routine cleaning activities, with a notably higher increase in Room A. This secondary increase was likely due to the resuspension of settled dust, rather than the duct cleaning itself.

Microbiological air samples were collected before and during duct cleaning in both rooms. In Room A, an increase in fungal spores was observed: A.fumigatus spore counts rose from 2 CFU/m³ before cleaning to 6 CFU/m³ during cleaning, while A. nidulans increased from 1 CFU/m³ to 4 CFU/m³. In contrast, Room B showed only minimal changes in fungal spore concentrations. A. fumigatus increased slightly from 1 CFU/m³ before cleaning to 2 CFU/m³ during cleaning, and A. nidulans increased from 0 to 1 CFU/m³ during cleaning.

Discussion

This study aimed to evaluate the impact of two different ventilation duct cleaning methods on air quality in a hospital setting. Our results demonstrate that internal duct cleaning using a standard cleaning method (Method A) led to a significant increase in airborne particle concentration and fungal spore release, whereas external duct cleaning using the novel 95G brushing method (Method B) had minimal impact on air quality. Particle counts in Room A peaked during cleaning and required approximately one hour to return to baseline levels, while microbiological air sampling showed increased concentrations of A. fumigatus and A. nidulans during in-room cleaning. In contrast, external cleaning did not significantly alter airborne particle levels or fungal spore counts in either room. Furthermore, our study suggests that routine room cleaning itself may contribute to increased particle dispersion, highlighting an additional potential source of contamination.

These findings emphasize the critical role of ventilation system maintenance in infection prevention and control (IPC), particularly in high-risk hospital areas. Hospital ventilation systems are essential for maintaining air quality, reducing airborne transmission of pathogens, and ensuring a safe environment for patients. The COVID-19 pandemic further underscored the importance of proper ventilation, as airborne pathogens, such as SARS-CoV-2, have been documented to spread via air conditioning systems, leading to outbreaks [6]. Our study supports existing concerns that ventilation systems can harbor and disperse microorganisms, particularly fungi, when not properly maintained.

Previous research has documented the presence and growth of microorganisms, particularly fungi such as Aspergillus spp., within ventilation systems. The extent of fungal growth is strongly influenced by the materials and construction of the air handling system, as well as key environmental conditions such as temperature, airflow velocity, and relative humidity [7]. However, to our knowledge, this is the first study examining the impact of duct cleaning on air quality in healthcare environments.

Given these risks, poorly maintained ventilation systems, where dust accumulation fosters fungal spore growth, pose a potential hazard to immunocompromised patients. A systematic review by Vonberg & Gastmeier [1] identified air supply systems as a significant source of nosocomial Aspergillus spp. outbreaks, contributing to 17% of reported cases, with airborne transmission route being the predominant route. Our findings are consistent with this concern, reinforcing that airborne contamination risk is not only linked to inadequate ventilation maintenance but also to the cleaning process itself.

While routine ventilation duct cleaning is recommended by the Centers for Disease Control and Prevention (CDC) [3] as part of HVAC maintenance, existing guidelines remain non-specific regarding dust cleaning methods. The CDC's Category IC recommendation only explicitly advises cleaning air-duct grilles when rooms are unoccupied to prevent dust accumulation. Our findings highlight the need for more detailed, evidence-based IPC policies to mitigate airborne contamination risks associated with ventilation maintenance in high-risk hospital units.

However, ventilation duct cleaning itself presents challenges. The process releases accumulated dust and fungal spores which may increase exposure risks for patients if not properly controlled. This necessitates organizational adjustments that can be difficult to implement in high-occupancy hospital wards. Our findings suggest that while duct cleaning during patient occupancy is feasible, in-room duct access significantly increases contamination risks, particularly for immunocompromised patients. External duct cleaning, by contrast, appears to be a safer alternative with minimal impact on air quality.

Additionally, our results suggest that routine room cleaning procedures contribute to air contamination. Following duct cleaning, particle counts increased during routine room cleaning, particularly in Room A, where the initial dust resuspension was highest. This underscores the challenges of maintaining clean air in healthcare facilities and highlights the need for further research on environmental cleaning methods and their impact on airborne contamination.

The main limitation of this study is its small scale. It was not designed as a large-scale research project, but rather as a practical assessment to address an immediate ventilation maintenance challenge. A larger, controlled study with reproducible results would be beneficial to validate these findings. Additionally, each healthcare facility is unique, and the generalizability of our method should be approached with caution. Furthermore, the lack of statistical rigor means that this study can only be considered as indicative rather than scientific evidence. Replication of this study in different hospital settings is strongly recommended before widespread implementation.

Conclusions

This study highlights the risks associated with internal duct cleaning, particularly for patients at heightened risk of invasive aspergillosis. External duct cleaning proved to be a safer alternative, minimizing the disruption to critical air quality in patient care areas. Our findings support the necessity of strategic planning in hospital ventilation maintenance to protect vulnerable patient populations from unnecessary airborne exposure.

Patient consent for publication

Not required.

Ethics approval

Not required.

Author contributions

Pascal De Waegemaeker (PDW), Thomas Snoeij (TS), Isabel Leroux-Roels (ILR).

The authors confirm that the authors have participated sufficiently in the work to take public responsibility for appropriate portions of the context and the manuscript has been red and approved by all named authors and that there are no others persons who satisfied the criteria for authorship but are not listed. The order of the authors listed in the manuscript has been approved by all the authors. PDW and TS conceived and designed the study; PDW performed data analysis; PDW drafted the manuscript; TS and ILR contributed to and revised the manuscript; all authors were involved in final approval of the version to be published.

Funding

No funding was involved.

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.