Journal of Hospital Infection 115 (2021) 117e123

Available online at www.sciencedirect.com

Journal of Hospital Infection

journal homepage: www.elsevier.com/locate/jhin

Laminar airflow versus turbulent airflow in simulated

total hip arthroplasty: measurements of colony-forming
units, particles, and energy consumption
L.V. Marsault a, C. Ravn a, b, A. Overgaard c, d, L.H. Frich e, M. Olsen f,
T. Anstensrud g, J. Nielsen h, S. Overgaard a, e, i, j, *
a
Odense University Hospital, Department of Orthopaedic Surgery and Traumatology, Odense, Denmark
b
Kolding Hospital Lillebaelt, Department of Orthopaedic Surgery and Traumatology, Kolding, Denmark
c
GentofteeHerlev Hospital, Department of Orthopaedic Surgery and Traumatology, Hellerup, Denmark
d
The Parker Institute, Bispebjerg and Frederiksberg Hospital, University of Copenhagen, Copenhagen, Denmark
e
University of Southern Denmark, Department of Clinical Research, Odense, Denmark
f
JRV A/S, Greve, Denmark
g
Niras, Odense, Denmark
h
Fournais Energy ApS, Vedbæk, Denmark
i
Copenhagen University Hospital, Bispebjerg, Department of Orthopaedic Surgery and Traumatology, Copenhagen
j
University of Copenhagen, Department of Clinical Medicine, Faculty of Health and Medical Sciences, Denmark

A R T I C L E I N F O S U M M A R Y

Article history: Background: The optimal type of ventilation in operating theatres for joint arthroplasty
Received 29 March 2021 has been debated for decades. Recently, the World Health Organization changed its
Accepted 20 June 2021 recommendations based on articles that have since been criticized. The economic and
Available online 25 June 2021 environmental impact of ventilation is also currently an important research topic but has
not been well investigated.
Keywords: Aim: To compare how large, high-volume, laminar airflow (LAF) and turbulent airflow
Ventilation (TAF) ventilation systems perform during standardized simulated total hip arthroplasty
Laminar airflow (THA), as they pertain to colony-forming units (cfu), particle counts, and energy
Turbulent airflow consumption.
Total hip arthroplasty Methods: Two identical operating theatres were used to perform simulated THA. The only
Energy consumption difference was that one was equipped with LAF and the other with TAF. Cfu and particles
Particle count were collected from key points in the operating theatre, and energy was measured for
each simulation. Thirty-two simulations were done in total.
Findings: LAF had significantly reduced cfu and particle count when compared with TAF,
at both 100% and 50% air influx. Furthermore, it was shown that lowering the air influx by
50% in LAF did not significantly affect cfu or particles, although reducing the fresh air
influx from 100% to 50% significantly lowered the energy consumption. Most simulations in
TAF did not meet the cleanroom requirements.
Conclusion: Cfu were significantly lower in LAF at both 100% and 50% air influx. It is
possible to reduce fresh air influx in LAF operating theatres by 50%, significantly reducing

* Corresponding author. Address: Copenhagen University Hospital, Bispebjerg, Department of Orthopaedic Surgery and Traumatology, Nielsine
Nielsens Vej 5, Entrance 6, 2nd floor, Copenhagen, NV, 2400, Denmark. Tel.: þ45 2063 4079.
E-mail address: soeren.overggard@regionh.dk (S. Overgaard).

https://doi.org/10.1016/j.jhin.2021.06.009
0195-6701/ª 2021 The Authors. Published by Elsevier Ltd on behalf of The Healthcare Infection Society. This is an open access article
under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
118 L.V. Marsault et al. / Journal of Hospital Infection 115 (2021) 117e123
energy consumption, while still maintaining cfu and particle counts below the ISO clas-
sification threshold required for THA surgery.
ª 2021 The Authors. Published by Elsevier Ltd
on behalf of The Healthcare Infection Society. This is an open access article
under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Introduction Methods
One of the most feared complications of total hip arthro- Study design
plasty surgery (THA) is infection. The risk herein is surgical site
infection e some will lead to prosthetic joint infections (PJIs), The mock-up operations were done on Fridays from October
which are associated with high morbidity, mortality, and cost 31st, 2014 to January 30th, 2015. There were three simulations
[1,2]. Many factors influence the risk of perioperative infec- each day, following a preconceived schedule to ensure
tion, of which patient comorbidity, perioperative antibiotics, changing ventilation between TAF and LAF as well as between
surgical hygiene, surgical technique, and ventilation are 100% and 50% fresh air within each ventilation (Table I).
important [3]. Thirty-two THA simulations were completed in 11 research
The optimal type of ventilation in operating theatres is an days, as planned. There were eight simulations with each of
ongoing discussion concerning the reduction of risk of infec- the following: laminar airflow at 100% fresh air influx (LAF100),
tion. Whether deep infections are related to airborne con- LAF at 50% fresh air influx (LAF50), turbulent airflow with 100%
tamination is still being debated [4e8]. fresh air influx (TAF100), and TAF with 50% fresh air influx
In 1982, Lidwell et al. recommended ultraclean ventilation, (TAF50).
but this was later questioned, leading to the current World The primary outcomes are a comparison of particle counts,
Health Organization (WHO) guidelines that discourage the use cfu, and energy consumption between LAF and TAF at 100%
of laminar airflow (LAF) systems, based on scarce evidence fresh air influx. Secondary outcomes are a comparison of the
[5,6,9,10]. This recommendation could be potentially harmful same parameters between LAF100 and LAF50 as well as between
to patients. TAF100 and TAF50, respectively. Lastly, LAF50 is compared to
However, it is ultimately very unlikely that randomized TAF100.
control trials will be done on the subject, as it would require This mock-up study was done according to a strict protocol
many participants to account for all the different hygienic (Appendix) for cleaning, placement of equipment in the ven-
measures taken during surgery. It would also require a long tilation field, patient placement, coverings, staffing, uniforms,
follow-up, due to PJIs arising up to two years after primary instruments, and movement patterns. In each simulation,
surgery. Accounting for these two factors would result in very there were five people with set roles: surgeon, assistant, scrub
costly research. nurse, nurse anaesthetist, and circulating nurse. A full-size,
Nonetheless, there are air quality requirements for operat- warmed-up mannequin was used, as well as the typical
ing theatres. European standards are described by the Inter- instruments and procedures used in THA surgery. Each simu-
national Standards Organisation (ISO) [11], as well as national lation lasted 50 min, which is the average time of the
guidelines, such as National Institute for Health and Care cementless THA most often used worldwide [21].
Excellence guidelines [12] and the British Orthopaedic Associ- The simulations were done in two identical full-size oper-
ation’s consultant advisory book [13]. Recommendations about ating theatres routinely used at Gentofte Hospital’s Depart-
particle load also exist. Particles >5 mm are correlated with ment of Orthopaedic Surgery. One was equipped with a large,
colony-forming units (cfu) and should therefore be minimized high-volume LAF system and the other with TAF ventilation.
[14e16]. Both operating theatres measure 44 m2, with a room volume
Ventilation units are expensive to install, and operating of 119 m3. The LAF ventilation covers 3.4  3.1 m and uses two
theatres account for a large part of the energy consumption in high-efficiency particulate air (HEPA) filters. The LAF system
hospitals. Most of the energy used comes from airflow run by
ventilation system fans, as well as cooling of the systems and Table I
heating the room air. Another reason why hospitals should be Simulation schedule
concerned about energy consumption is the environment. The
Simulation day Date Type of ventilation
European Commission has launched the green@hospital ini-
tiative, to fund research in the direction of climate-friendly 1 Oct 31st, 2014 LAF100  3
hospitals [17]. There is minimal evidence on the effect of 2 Nov 7th, 2014 TAF100  3
lowering the airflow e and thus energy consumption in oper- 3 Nov 14th, 2014 LAF50  3
ating theatres e due to hygienic concerns outweighing the 4 Nov 21st, 2014 TAF50  3
potential energy savings [18e20]. As far as we know, no 5 Nov 28th, 2014 LAF100  3
studies have compared energy consumption between turbu- 6 Dec 5th, 2014 TAF100  3
lent airflow (TAF) and LAF when lowering air influx in simu- 7 Dec 19th, 2014 LAF50  3
lated THA. 8 Jan 9th, 2015 TAF50  3
Therefore, this study aims to compare LAF with TAF ven- 9 Jan 16th, 2015 LAF100  2 TAF100  1
tilation regarding cfu, particle counts, and energy consumption 10 Jan 23rd, 2015 TAF100  1 TAF50  2
during simulated THA surgery at 100% vs 50% fresh air influx. 11 Jan 30th, 2015 LAF50  2
 L.V. Marsault et al. / Journal of Hospital Infection 115 (2021) 117e123 119

1 = Surgeon

4
2 = Scrub nurse

2
3 = Assistant

1
4 = Circulating nurse

3
5 = Nurse anaesthetist = Operating particle counter

= Air sampler
P1 P2
= Agar plate
5 P1 = Particle counter
P2 = Particle and cfu counter
= Particle collector
= Camera

LAF

1 = Surgeon

= Scrub nurse
4

2
2

3 = Assistant

= Circulating nurse
1

4
3

5 = Nurse anaesthetist = Operating particle counter

= Air sampler
P1 P2
= Agar plate
5 P1 = Particle counter
P2 = Particle and cfu counter
= Particle collector
= Camera

TAF

Figure 1. Simulation set-up in each operating theatre. LAF, laminar airflow; TAF, turbulent airflow.

circulates the air, mixing recycled air from the operating the-
atre with fresh air from the outside. The airflow rate of fresh
air is 2760 m3/h, and recirculated air is 7075 m3/h. The TAF
ventilation uses one HEPA filter identical to the LAF system and
has an airflow rate of 2533 m3/h, all of which is outside air.
For the simulations with reduced air influx, the hospital’s
building management system was used to reduce fresh air
influx to 50%. This increases the amount of air that is recircu-
lated, due to the mixing chamber being situated on the suction
side of the LAF fans. Thus, the influx of air through the LAF unit
is unaltered for both situations (100%, 50%). The overpressure
in the operating theatre is regulated by the VAV terminal
(variable air volume damper unit) in the extraction duct
system. Reduction of the outdoor air leads the control system
to reduce the extraction from the operating theatre to keep
the set-point value for the overpressure value.
Microbial sampling
In each simulation, four agar plates (5% blood agar, diame-
ter 9 cm; Statens Serum Institut, Copenhagen, Denmark) were
used according to the protocol (Figure 1) in passive and active
sampling, respectively. On the morning of each simulation day,
a baseline measurement was made in the centre of the ven-
tilation zone with active sampling before any other access to
120 L.V. Marsault et al. / Journal of Hospital Infection 115 (2021) 117e123

Table II
Recommended threshold for cfu and particle counts in ultra cleanrooms
Cleanroom status Active cfu/m3 Passive cfu/4 h (plate diameter: 90 mm) Particles (particles/m3)
Large (5e10 mm) All (0.5e10 mm)
At rest 0.029  10 3
3.52  103
In use 10 5 2.9  103 352  103
cfu, colony-forming unit.
ISO Class 5 (EU grade B) ventilation is recommended for total hip arthroplasty surgery.

the operating theatre. One unexposed plate was analysed in
the same way and served as the negative control.
Active sampling was done with a calibrated Microbiological
Active Sampler (MAS-100 NT; Merck KGaA, Darmstadt, Ger-
many). The samplings were done in 5 min intervals, with a flow
rate of 100 L/min, thereby sampling 1.0 m3 air.
During each simulation, the sampler was placed 30 cm from
the surgical field. The first sampling was done at the first
incision and the second sample during the sawing of the fem-
oral neck. In the results, the active samples are summed, and
data shown in cfu/m3.
Passive sampling was performed with the two remaining
plates, one being placed next to the surgical field and the other
on the instrument table closest to the edge of the ventilation
zone. Both plates were exposed during the entire surgery of 50
min. Passive sedimentation on to the agar plates was measured
in cfu/m2/h.
All plates were incubated at 35 C for about two days. Cfu
were then counted for each plate. Matrix-assisted laser
desorptioneionization mass spectrometry (Bruker Biotyper,
Bremen, Germany) was used to differentiate bacteria strains.
The thresholds used in the present study (Table II) were
chosen according to the European guidelines [13,22]. They
recommend that air should be ultraclean during arthroplasty
surgeries, regardless of the type of ventilation.
Particle count
Sampling was done three times:
1. For 5 min on the morning of simulation days, before other
access.
2. Before each simulation, a 5 min baseline sample after set-
up, while personnel stood still.
3. During each simulation, a continuous 50 min measurement.
Previous studies using particle size to describe air con-
tamination in operating theatres found a correlation between
large particles (5 mm) and cfu, whereas the smaller particles
do not correlate with cfu [15,16,23e26]. Therefore, using the
same grouping as the ISO, only sizes 0.5e10 mm are accounted
for [11]. Results are shown as particles/m3 air for all particles
(0.5e10 mm) and large particles (5e10 mm).
Thresholds for air quality regarding airborne particles in
cleanroom ventilation are regulated by ISO 14644-1 [11,27,28].
This study used the thresholds for Class 5 cleanrooms (Table II).
Energy consumption
A Bender PEM735 Power Analyzer was mounted on the
electrical supply side for the ventilation system fans. The
energy consumption in kWh displayed was noted at the begin-
ning and end of each simulation. Energy consumption for
warming/cooling coils was collected from the hospital’s
building management system.

The particle count was measured with a portable airborne Statistics

particle counter (MET one, model 3413; Beckman Coulter, CA,
USA). It uses a laser to count particles of standard sizes: 0.3,
0.5, 1.0, 3.0, 5.0, and 10.0 mm. The particles in our study were
gathered with active suction (flow rate 28.3 L/min) through a
funnel probe beside the surgical field and led to the particle
counter through a plastic tube w2 m long.
Data were logged and calculated in Microsoft Excel using the
WinSTAT add-on (R. Fitch Software, Cambridge, MA, USA).
Results are shown as median and range. The ManneWhitney U-
test was used for all comparisons. P < 0.05 was considered
statistically significant.

Table III
Cfu sampling during simulated THA surgery in theatres with LAF and TAF using 50% and 100% ventilation
Sampling 100% fresh air 50% fresh air
LAF TAF LAF TAF
3
Active sampling (cfu/m )
Surgical site 0 (0e4)a,b 15.5 (11e21)c,d 1 (0e2)a,b 12.5 (5e44)c,d
Passive sampling (cfu/m2/h)
Surgical site 0 (0)a,b 0.5 (0e3)c,d 0 (0)a,b 1.5 (0e8)c,d
Instrument table 0 (0e1)b 0.5 (0e2) 0 (0e2) 1.5 (0e4)c
cfu, colony-forming unit; THA, total hip arthroplasty; LAF, laminar airflow; TAF, turbulent airflow.
Values are median (range); N ¼ 8.
a
P < 0.05 vs TAF100.
b
P < 0.05 vs TAF50.
c
P < 0.05 vs LAF100.
d
P < 0.05 vs LAF50.
 L.V. Marsault et al. / Journal of Hospital Infection 115 (2021) 117e123 121

Table IV
Particle sampling during simulated THA surgery in theatres with LAF and TAF using 50% and 100% ventilation
Particle type 100% fresh air 50% fresh air
LAF TAF LAF TAF
Large (5e10 mm)/m  10
3 3
0.6 (0.3e12) a,b
11 (3.8e24) c,d
1.3 (0.1e8.8) a,b
8.0 (4.5e15)c,d
All (0.5e10 mm)/m3  103 21 (8.2e125)a,b 224 (82e502)c,d 25 (3.2e106)a,b 174 (91e326)c,d
THA, total hip arthroplasty; LAF, laminar airflow; TAF, turbulent airflow.
Median (range); N ¼ 8.
a
P < 0.05 vs TAF100.
b
P < 0.05 vs TAF50.
c
P < 0.05 vs LAF100.
d
P < 0.05 vs LAF50.

Results
Microbial sampling
Control measurements
None of the negative control plates had any growth. The
active morning samples before simulation showed no growth in
LAF100, whereas LAF50 had a median of 2 cfu (range: 0e3).
TAF100 had a median of 4.5 cfu (range: 1e17) and TAF50 12 cfu
(range: 5e22).
Measurements during simulation
In every simulation in LAF-ventilated operating theatres,
the cfu count during active sampling was under the recom-
mended limit. However, TAF exceeded this threshold in 14/16
experiments. For passive measurements, TAF did not exceed
the threshold.
In decreasing order, bacterial strains found include
M. luteus, S. epidermidis, and S. warneri, Moraxella sp. and
Bacillus sp.
LAF100 had significantly fewer actively sampled cfu than
TAF100 (P ¼ 0.0005) (Table III). There was no significant dif-
ference between LAF100 and LAF50 (P ¼ 0.4) or between TAF100
and TAF50 (P ¼ 0.9). However, LAF50 had significantly fewer cfu
than TAF100 (P ¼ 0.0007).
Particle count
Control measurements
The morning measurements in the LAF operating theatre all
showed zero particles of any size. In the TAF operating theatre,
morning measurements showed a median all-size particle
count of 562 and 534 particles/m3 (range: 161e1738) for TAF100
and TAF50, respectively, with no significant difference between
the two (P ¼ 0.8). That is below the threshold for cleanroom
ventilation at rest. However, the morning measurements of
large particles were only below the threshold in one of six
simulation days where the TAF operating theatre was used. The
median was 38.5 and 39 particles/m3 (range: 14e79) with no
difference between TAF100 and TAF50 (P ¼ 0.8).
Measurements during simulation
In the LAF operating theatre, the threshold for large par-
ticles was exceeded in four out of 16 experiments. The
threshold for all-size particles was never exceeded. In the TAF
operating theatre, the threshold for large particles was
exceeded by 28e700% in every simulation. The threshold for
all-size particles was exceeded once, by 42%, with TAF100
ventilation.
LAF100 had significantly lower median particle counts of
both the large and all-size particle counts compared to TAF100
(P ¼ 0.005 and P ¼ 0.002, respectively) (Table IV). There was
neither a difference between LAF100 and LAF50 (P ¼ 0.9 and P ¼
0.7) nor between TAF100 and TAF50 (P ¼ 0.7 and P ¼ 1). How-
ever, LAF50 had lower particle counts than TAF100 (P ¼ 0.005
and P ¼ 0.003).
Energy
When decreasing air influx from 100% to 50% in the LAF
operating theatre, energy consumption was reduced by 41%
(P ¼ 0.0007) (Table V). In the TAF operating theatre, energy
consumption was reduced by 51% (P ¼ 0.0007) when reducing
air influx from 100% to 50%.
The energy consumption for TAF100 was significantly lower
than for LAF100 (P ¼ 0.001), whereas consumption in LAF50 was
lower than for TAF100 (P ¼ 0.0007).

Table V
Discussion
Energy consumption (kWh) during simulated surgery on theatres
The purpose of this study was to compare how large, high-
with LAF and TAF using 50% and 100% ventilation
volume LAF and TAF ventilation systems at 100% and 50%
Procedure 100% fresh air 50% fresh air fresh air influx perform during standardized simulated THA, as
LAF 1.85 (1.66e2.03)a,b,d 1.12 (0.95e1.31)a,b,c it pertains to bacteria and particle counts as well as energy
TAF 1.54 (1.53e1.83)b,c,d 0.75 (0.73e0.87)a,c,d consumption.
This study showed that large, high-volume LAF had sig-
LAF, laminar airflow; TAF, turbulent airflow.
Median (range); N ¼ 8. nificantly reduced bacterial and particle counts compared with
a
P < 0.05 vs TAF100. TAF. Furthermore, it was shown that lowering the fresh air
b
P < 0.05 vs TAF50. influx to 50% in LAF did not significantly affect the bacterial or
c particle counts, while significantly lowering the energy
P < 0.05 vs LAF100.
d
P < 0.05 vs LAF50. consumption.
122 L.V. Marsault et al. / Journal of Hospital Infection 115 (2021) 117e123
One strength of this mock-up study is that we had unique
standardized simulated surgeries, which is not possible clin-
ically. Moreover, every case was performed as in real-time THA
surgery using a standardized protocol. This includes stand-
ardized operation time, surgical procedure, surgical attire, and
the number of staff, among others. A simulated controlled set-
up is important when specific factors are studied. The simu-
lations were all done by the same surgeon, limiting inter-
operator differences in surgical technique. With a simulation,
every aspect of surgery could be controlled, including patient
parameters, and the significance of ventilation alone could be
evaluated.
Limitations to the study include using a heated dummy
instead of a patient. Thus, it was not possible to use a cautery
instrument, and there was no bleeding or moisture. Fur-
thermore, we found no S. aureus strains in the present study,
even though it is the most common strain found in PJI [29e30].
S. aureus mainly originates from skin flora, which the dummy
did not have. However, we do not consider this a major limi-
tation to the present study, as ventilation should not impact
auto-contamination from the patient, and would therefore not
change our findings when comparing ventilation type. Another
limitation is having only performed simulations in the winter
months, where the energy consumption may differ from the
rest of the year. However, alternating all the simulations in
that same period makes the comparison valid. Lastly, we used
cfu/m3 as a surrogate endpoint for PJI, even though not every
airborne bacterium produces PJI. Despite surrogate endpoints
being a limitation, cfu is notably the measurement used to
classify cleanrooms.
Only a few studies have measured cfu during surgery or
simulated surgery, but these have not compared TAF and LAF
operating theatres during simulated THA surgery [19e20].
Alsved et al. showed that using a temperature-controlled air-
flow system (TcAF) to obtain laminar airflow also performs
better than conventional TAF [20]. LAF performed better than
TcAF, although both remained in the same cleanroom class.
Alsved et al. found that TcAF uses less energy than LAF, while
still creating the protective airflow around the surgical table.
However, they did not measure energy directly during surgery
but had it calculated by an external consultation agency. Their
study is limited by not having the control given by simulation
and having a mix of surgeries on different joints.
A recent study of the Norwegian arthroplasty register
showed that THA surgeries performed in validated, large, high-
volume LAF systems had a 20% lower risk of revision due to
infection compared to TAF [31]. It is important to note, how-
ever, that Langvatn et al. evaluated the risk of revision due to
infection, whereas we measured airborne bacteria as a surro-
gate endpoint for the risk of PJI [31]. However, Langvatn et al.
concluded that the perception of LAF systems being potentially
harmful compared to TAF seems erroneous [5,6]. This is partly
due to the scientific basis of such conclusions seeming to rest
primarily on studies from surveillance registries. Those studies
assess LAF as one big entity and do not account for the
potential misreporting, as was also demonstrated by Langvatn
[32]. The uncertain validity due to poorly validated ventilation
data and potential misreporting in the studies on which WHO
based their recommendations leads to concerns that these
recommendations are potentially harmful [10].
We found that TAF ventilation does not always live up to the
ISO Class 5 standards. This is in line with previous studies
[19e20]. Our study showed that TAF ventilation exceeded the
standards for large particles in every simulation, regardless of
ventilation setting. The cfu threshold of 10/m3 was exceeded
in 14 out of 16 simulations during active sampling in the TAF
operating theatre. This is a great concern as several new hos-
pital buildings may have installed TAF rather than LAF in the
operating theatres. By contrast, LAF had values within the ISO
Class in all procedures.
Energy consumption in hospitals has been debated, as made
evident by ongoing studies financed by the European Commis-
sion’s Green@Hospital initiative [17]. Consequently, we inves-
tigated the effect of lower air influx and thereby energy
consumption. It was found that even when decreasing LAF fresh
air influx by 50% and maintaining room pressure, LAF still met
the criteria for cleanroom. These state that large and all par-
ticles should respectively be below 2900 and 352,000 particles/
m3 while in use, and cfu should be <10/m3. This is in line with
Gormley et al.’s findings that increasing air changes did not
improve air quality, whereas decreasing ventilation rate
decreased air quality, although not enough to change the ISO
classification [19]. This suggests that a minimum airflow is
required to maintain air quality but that higher flow may not
necessarily be better. We found that LAF at 50% performs
better than TAF at 100%, regarding both cfu and particles while
also using less energy. It may therefore be worth researching
further to find the right balance between air quality and cost.
This could potentially be achieved with in-house real-time air
quality measurements that, in turn, could be set up to auto-
matically adjust ventilation rates within a set range.
In conclusion, this study shows that LAF had significantly
reduced bacterial and particle counts compared with TAF.
Furthermore, it shows that lowering the fresh air influx by 50%
in LAF did not significantly affect bacterial or particle counts,
while significantly lowering the energy consumption.
Furthermore, the TAF ventilation exceeded the recom-
mended thresholds for cfu and particles in the majority of
simulations. In perspective, we suggest that the current
guidelines allowing THA surgery in TAF ventilation should be re-
evaluated [10]. There is a need for green solutions in the
future, and reduction of the fresh air influx in LAF operation
theatres could potentially be one.
Acknowledgements
We would like to especially thank the OP-Vent project group
and the Department of Orthopaedic Surgery at GentofteeHerlev
hospital, as well as E. Knudsen and M. Kemp from the Department
of Clinical Microbiology at Odense University Hospital.
Conflict of interest statement
None declared.
Funding source
This study was financed by Elforsk grant number 345-019.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jhin.2021.06.009.
 L.V. Marsault et al. / Journal of Hospital Infection 115 (2021) 117e123
References
[1] Vanhegan IS, Malik AK, Jayakumar P, Ul Islam S, Haddad FS.
A financial analysis of revision hip arthroplasty. J Bone Joint
Surg Br 2012;94-B:619e23. https://doi.org/10.1302/0301-620X.
94B5.27073.
[2] Gundtoft PH, Pedersen AB, Varnum C, Overgaard S. Increased
mortality after prosthetic joint infection in primary THA. Clin
Orthop Relat Res 2017;475:2623e31. https://doi.org/10.1007/
s11999-017-5289-6.
[3] Alamanda VK, Springer BD. Perioperative and modifiable risk
factors for periprosthetic joint infections (PJI) and recommended
guidelines. Curr Rev Musculoskelet Med 2018;11:325e31. https://
doi.org/10.1007/s12178-018-9494-z.
[4] Popp W, Alefelder C, Bauer S, Daeschlein G, Geistberger P,
Gleich S, et al. Air quality in the operating room: surgical site
infections, HVAC systems and discipline e position paper of the
German Society of Hospital Hygiene (DGKH). GMS Hyg Infect
Control 2019;14:Doc20. https://doi.org/10.3205/dgkh000335.
[5] Gastmeier P, Breier A-C, Brandt C. Influence of laminar airflow on
prosthetic joint infections: a systematic review. J Hosp Infect
2012;81:73e8. https://doi.org/10.1016/j.jhin.2012.04.008.
[6] Bischoff P, Kubilay NZ, Allegranzi B, Egger M, Gastmeier P. Effect of
laminar airflow ventilation on surgical site infections: a systematic
review and meta-analysis. Lancet Infect Dis 2017;17:553e61.
https://doi.org/10.1016/S1473-3099(17)30059-2.
[7] Chauveaux D. Preventing surgical-site infections: measures other
than antibiotics. Orthop Traumatol Surg Res 2015;101:S77e83.
https://doi.org/10.1016/j.otsr.2014.07.028.
[8] Salassa TE, Swiontkowski MF. Surgical attire and the operating
room: role in infection prevention. J Bone Joint Surg Am
2014;96:1485e92. https://doi.org/10.2106/JBJS.M.01133.
[9] Lidwell OM, Lowbury EJ, Whyte W, Blowers R, Stanley SJ,
Lowe D. Effect of ultraclean air in operating rooms on deep
sepsis in the joint after total hip or knee replacement: a
randomised study. BMJ 1982;285:10e14. https://doi.org/10.1
136/bmj.285.6334.10.
[10] World Health Organization. Global guidelines for the prevention
of surgical site infection. Geneva: WHO; 2016.
[11] ISO 14644-1:2015. Cleanrooms and associated controlled environ-
ments e Part 1: Classification of air cleanliness by particle con-
centration. Available at: https://www.iso.org/standard/53394.
html [last accessed May 2020].
[12] NICE Recommendations. Joint replacement (primary): hip, knee
and shoulder. Guidance. Available at: https://www.nice.org.uk/
guidance/ng157/chapter/Recommendations [last accessed June
2021].
[13] British Orthopaedic Association. Consultant advisory book. n.d.
Available at: https://www.boa.ac.uk/standards-guidance/
consultant-advisory-book.html [last accessed August 2020].
[14] Hansen D, Krabs C, Benner D, Brauksiepe A, Popp W. Laminar air
flow provides high air quality in the operating field even during real
operating conditions, but personal protection seems to be neces-
sary in operations with tissue combustion. Int J Hyg Environ Health
2005;208:455e60. https://doi.org/10.1016/j.ijheh.2005.08.008.
[15] Stocks GW, Self SD, Thompson B, Adame XA, O’Connor DP. Pre-
dicting bacterial populations based on airborne particulates: a
study performed in nonlaminar flow operating rooms during joint
arthroplasty surgery. Am J Infect Control 2010;38:199e204.
https://doi.org/10.1016/j.ajic.2009.07.006.
[16] Seal DV, Clark RP. Electronic particle counting for evaluating the
quality of air in operating theatres: a potential basis for stand-
ards? J Appl Bacteriol 1990;68:225e30. https://doi.org/10.1111/
j.1365-2672.1990.tb02568.x.
[17] European Commission. Green@Hospital j Making hospitals
“healthier” and greener. Shaping Europe’s digital future e
European Commission 2015. Available at: https://ec.europa.eu/
123
digital-single-market/en/content/greenhospital-making-
hospitals-healthier-and-greener [last accessed September 2020].
[18] Zoon WAC, Heijkant SAM, Hensen J, Loomans MGLC. Assessment of
the performance of the airflow in an operating theatre. Helsinki:
SCANVAC Roomvent; 2007. Available at: https://www.researchgat
e.net/publication/254842907_Assessment_of_the_performance_
of_the_airflow_in_an_operating_theatre [last accessed May 2020].
[19] Gormley T, Markel TA, Jones H, Greeley D, Ostojic J, Clarke JH,
et al. Cost-benefit analysis of different air change rates in an
operating room environment. Am J Infect Control 2017;45:1318e23.
https://doi.org/10.1016/j.ajic.2017.07.024.
[20] Alsved M, Civilis A, Ekolind P, Tammelin A, Andersson AE,
Jakobsson J, et al. Temperature-controlled airflow ventilation in
operating rooms compared with laminar airflow and turbulent
mixed airflow. J Hosp Infect 2018;98:181e90. https://doi.org/
10.1016/j.jhin.2017.10.013.
[21] Gundtoft PH, Pedersen AB, Schønheyder HC, Møller JK, Overgaard S.
One-year incidence of prosthetic joint infection in total hip arthro-
plasty: a cohort study with linkage of the Danish Hip Arthroplasty
Register and Danish Microbiology Databases. Osteoarthr Cartil
2017;25:685e93. https://doi.org/10.1016/j.joca.2016.12.010.
[22] ISO 14698-1:2003. Cleanrooms and associated controlled envi-
ronments e Biocontamination control e Part 1: General princi-
ples and methods. Available at: https://www.iso.org/cms/
render/live/en/sites/isoorg/contents/data/standard/02/50/
25015.html [last accessed August 2020].
[23] Friberg B, Friberg S, Burman LG. Correlation between surface and
air counts of particles carrying aerobic bacteria in operating
rooms with turbulent ventilation: an experimental study. J Hosp
Infect 1999;42:61e8. https://doi.org/10.1053/jhin.1998.0542.
[24] Scaltriti S, Cencetti S, Rovesti S, Marchesi I, Bargellini A, Borella P.
Risk factors for particulate and microbial contamination of air in
operating theatres. J Hosp Infect 2007;66:320e6. https://doi.org/
10.1016/j.jhin.2007.05.019.
[25] Nilsson K-G, Lundholm R, Friberg S. Assessment of horizontal
laminar air flow instrument table for additional ultraclean space
during surgery. J Hosp Infect 2010;76:243e6. https://doi.org/
10.1016/j.jhin.2010.05.016.
[26] Cristina ML, Spagnolo AM, Sartini M, Panatto D, Gasparini R,
Orlando P, et al. Can particulate air sampling predict microbial
load in operating theatres for arthroplasty? PLoS One
2012;7:e52809. https://doi.org/10.1371/journal.pone.0052809.
[27] Dharan S, Pittet D. Environmental controls in operating theatres.
J Hosp Infect 2002;51:79e84. https://doi.org/10.1053/
jhin.2002.1217.
[28] US Food and Drug Administration and EU Good Manufacturing
Processes. Annex 1. Differences in cleanroom specifications. n.d.
Available at: https://www.raps.org/news-and-articles/news-
articles/2019/7/fda-and-eu-gmp-annex-1-differences-in-
cleanroom-sp [last accessed May 2020].
[29] Rakow A, Perka C, Trampuz A, Renz N. Origin and characteristics of
haematogenous periprosthetic joint infection. Clin Microbiol Infect
2019;25:845e50. https://doi.org/10.1016/j.cmi.2018.10.010.
[30] Tsai Y, Chang C-H, Lin Y-C, Lee S-H, Hsieh P-H, Chang Y. Different
microbiological profiles between hip and knee prosthetic joint
infection. J Orthop Surg (Hong Kong) 2019 MayeAug;27(2).
https://doi.org/10.1177/2309499019847768. 2309499019847768.
[31] Langvatn H, Schrama JC, Cao G, Hallan G, Furnes O, Lingaas E,
et al. Operating room ventilation and the risk of revision due to
infection after total hip arthroplasty: assessment of validated
data in the Norwegian Arthroplasty Register. J Hosp Infect
2020;105:216e24. https://doi.org/10.1016/j.jhin.2020.04.010.
[32] Langvatn H, Bartz-Johannessen C, Schrama JC, Hallan G,
Furnes O, Lingaas E, et al. Operating room ventilation e vali-
dation of reported data on 108,067 primary total hip arthro-
plasties in the Norwegian Arthroplasty Register. J Eval Clin Pract
2020;26:1022e9. https://doi.org/10.1111/jep.13271.