nursing

ORIGINAL ARTICLE • HIP - INFECTION

Does laminar airflow make a difference to the infection rates for lower limb arthroplasty: a study using the National Joint Registry and local surgical site infection data for two hospitals with and without laminar airflow

S. Singh1 • S. Reddy1 • Raj Shrivastava1

Received: 19 May 2016 / Accepted: 4 September 2016 / Published online: 29 September 2016

� Springer-Verlag France 2016

Abstract This study compared the National Joint Registry

for England, Wales, Northern Ireland and the Isle of Man

(NJR) data for total hip replacements (THRs) and total knee

replacements (TKRs) from Hospital A [with laminar airflow

(LAF)] and Hospital B (without LAF). These hospitals were

originally managed by two different trusts that subsequently

merged. Consequently, the theatres in Hospital A have

always had LAF and those in Hospital B had only conven-

tional ventilation systems. As this merger happened before

the establishment of the NJR, it puts us in a unique position,

enabling direct comparison of the revision rates for infected

hip and knee replacements between the two hospitals that

follow similar infection protocols. Analysis of the NJR data

showed there were no statistical differences. Of the 2234

TKRs performed at Hospital A, 16 were revised for infection,

whereas 19 of the 3694 TKRs at Hospital B were revised

(p \ 0.33). Of the 1752 THRs at Hospital A, 5 were revised for infection, whereas this was the case for 12 of the 3163

THRs at Hospital B (p \ 0.59). There was also no statistical difference when combining the figures for TKRs and THRs

(p \ 0.59). Our local surgical site infection (SSI) data from these two hospitals were also analysed. Again, there was no

statistical difference between the two sites (p \ 0.34). Using LAF has not reduced the rate of revision for infection nor it

has reduced the incidence of SSI in our theatres. This is the

first study comparing infection rates in two different hospi-

tals serving similar patient populations using the NJR and

SSI data. Our study questions the rationale of increasing use

of LAF in routine lower limb arthroplasty. We call for greater

debate and more robust studies on the subject.

Keywords Lower limb arthroplasty � Total knee replacement � Total hip replacement � Infection � Laminar air flow � Surgical site infection (SSI) � National joint registry (NJR)

Introduction

Deep infection is a dreaded complication of joint arthroplasty.

Although infrequent [1, 2], it represents a significant mor-

bidity for the patient and a huge cost to healthcare [3–5]. Since

the advent of joint arthroplasty, a multitude of factors have

beeninvestigatedtoreducethe infectionrates.Oneofthemost

commonly researched factors is the cleanliness of theatre air.

Charnley and other earlier researchers brought to atten-

tion the importance of having clean air in the operating

theatre [6–9]. This was based primarily on the concept of

controlling how often the air is changed per minute and its

direction. Clean air in operating theatres is achieved largely

by laminar airflow (LAF) systems and space suits [10].

Conventional plenum ventilation systems exchange the

theatre air about 30 times per hour, whereas the LAF theatres

exchange it over 300 times an hour. This should result in a

maximum of 10 colony-forming units per cubic metre (cfu/

m 3 ) with values sometimes reaching as low as 1 cfu/m

3 . The

Medical Research Council trial confirmed the importance of

LAF theatres in the reduction of colony-forming units [11].

More recently, there have been a number of articles

showing either no difference in the infection rates or even

In the original publication of this article, the family name of

corresponding author has been published incorrectly; this error has

now been corrected.

& Raj Shrivastava [email protected]

1 Department of Orthopaedics, William Harvey Hospital, East

Kent Hospitals University NHS Foundation Trust, Ashford,

Kent TN24 0LZ, UK

123

Eur J Orthop Surg Traumatol (2017) 27:261–265

DOI 10.1007/s00590-016-1852-1

slightly higher rates of infection with the use of LAF in

theatres [10, 12–15]. This goes against the accepted wis-

dom of current orthopaedic teaching and practice [16].

One of the challenges in investigating the effect of an

intervention on deep infection is that it can be multi-fac-

torial and infection rates remain low [1, 2]; this makes it

very difficult to set up randomised controlled trials with

adequate statistical power [10, 17]. Joint registries provide

reliable data on a large number of patients that are col-

lected prospectively. This produces powerful data for rare

complications such as infection [13].

We are in a unique position with two hospitals in the

same trust (and therefore serving similar population

demographics) both undertaking joint replacements rou-

tinely. The hospitals were originally managed by two dif-

ferent trusts, and consequently the theatres of Hospital A

have always had a LAF (ultraclean vertical laminar airflow

system, Medical Air Technology, Manchester, M41 7LY,

UK), whereas the theatres in Hospital B have always had

conventional ventilation systems.

Data from both hospitals have been routinely submitted to

the National Joint Registry (NJR) since its inception. As this

merger happened before the establishment of the NJR, it gave

us a unique opportunity to compare the revision rates for both

infected total hip replacements and total knee replacements

between the two hospitals that follow similar infection pro-

tocols. Our hypothesis was that the hospital with LAF theatres

would have lower rates of revision for infected joints.

Patients and methods

Data were obtained from the NJR for all hip and knee

replacements carried out at our two hospitals from April

2003 to October 2014. These data contained the numbers

revised for infective causes from the index groups.

In addition, data were reviewed for SSI in hip and knee

replacements performed between April 2008 and March

2013. These data were collected retrospectively by sifting

through case records, operative notes, discharge summaries

and microbiology swabs. For our hospitals, they are rou-

tinely verified by a consultant microbiologist and a team of

consultant orthopaedic surgeons at a weekly multi-

disciplinary team meeting. The SSI data do not differen-

tiate between hip and knee cases and therefore represent

infections for hip and knee replacements combined. These

data are for the infections presenting within 12 months of

the arthroplasty procedure and represent the early infec-

tions, these are the ones that are most likely to be influ-

enced by intra-operative factors like the LAF.

There were two outcome variables: revision due to

infection and SSI. Owing to the binary nature of the out-

comes, the difference between the two hospitals was

assessed using the Chi-square test. The analyses were

performed for all operations combined as well as separately

for hip and knee operations for the NJR dataset.

The hip and knee arthroplasty data from local operating

theatre lists were used to extract the information about the

patient demographics and the general health of the patients

as evidenced by their American Society of Anaesthesiol-

ogists (ASA) score. The analyses were performed using the

Mann–Whitney test for age and ASA score, and the Chi-

square test for gender.

A p value of less than 0.05 was considered statistically

significant in our study.

Results

The results for revision arthroplasty and SSIs in the two

hospitals are shown in Table 1, whereas, Tables 2 and 3

summarise the patient demographics and their ASA score.

There were no statistically significant differences

between the hospitals for revision rates or SSI rates.

There were statistically significant differences in ASA

score for both sets of data. The scores were generally lower

for Hospital A than Hospital B suggesting that patients in

Hospital B had more comorbidities. The age of hip patients

was also lower, on average, in Hospital A.

Discussion

Prevention of infection remains one of the most cherished

goals of joint replacement surgery. Various researchers

have investigated the possible causes of infection in

Table 1 Summary of number of revisions and SSIs in the two hospitals

Outcome Group Hospital A (LAF) Hospital B (non-LAF) Difference* (95% CI) p value

Revision (infection) Hips 5/1752 (0.29%) 12/3163 (0.38%) 0.09% (-0.24, 0.42%) 0.59

Knees 16/2234 (0.72%) 19/3674 (0.51%) -0.20% (-0.62, 0.22%) 0.33

Hips and knees 21/3986 (0.53%) 31/6837 (0.45%) -0.07% (-0.35, 0.20%) 0.59

SSI Hips and knees 31/3361 (0.92%) 56/4863 (1.14%) 0.22% (-0.22, 0.65%) 0.34

CI confidence interval

* Calculated as result for Hospital B minus result for Hospital A

262 Eur J Orthop Surg Traumatol (2017) 27:261–265

123

arthroplasty, including the role played by operating room

air as a source of infection. The need to improve the air

(mainly by special ventilation systems in the operating

theatres) has long been the subject of debate.

Charnley realised the importance of clean air in oper-

ating theatres and was able to achieve a 20-fold reduction

in his infection rates by using occlusive garments for

operating room personnel [6, 7]. Lidwell [11] also rec-

ommended the use of LAF in theatres although his study

was widely criticised because of variable and uncontrolled

use of peri-operative prophylactic antibiotics.

More recent publications by Hooper [10] and Gastmeier

[13] show no difference in the infection rates of procedures

performed in LAF and non-LAF theatres. Brandt et al. [18]

using the German KISS (Krankenhaus-Infektions-

Surveillance-System) data in 2008 concluded that operating

room ventilation with LAF provided no protective effect on

the SSI rates in orthopaedic and abdominal surgery. This

study was widely discussed in Germany with one of the

major discussion points was the hypothesis that only ceiling

distribution systems with a size of 3.2 m 9 3.2 m 2 or larger

work appropriately and this study did not consider the size

of LAF ceilings. The critics argued that ceiling distribution

size of lesser area would result in particle free protection

zones that are too small and lead to unwanted turbulence at

the fringe areas directly above the surgical team and the

surgical instruments outside the operation field [18, 19].

Consequently, Breier et al. [20] conducted a study looking

specifically at the ceiling distribution size in German the-

atres. This study confirmed that there was no difference in

the SSI rate, even when taking the distribution ceiling area

of LAF systems into consideration.

In another systematic review of the influence of LAF on

prosthetic joint infection, Gastmeier et al. [13] suggest that

not only there is no difference but there could be a chance

of increased infection in LAF theatres [10, 12, 20]. They

found that the two main explanations for the higher SSI

risk with LAF in literature were inappropriate positioning

of patients and personnel in the operating theatre, creating

air eddies [10, 21], and lower intra-operative tissue tem-

peratures in the surgical wound [18]. In fact, randomized

trials have shown that forced warming of the air in theatre

substantially decreases the risk of SSI, and guidelines with

regards to the importance of maintaining peri-operative

normothermia for the prevention of SSI have been pub-

lished [22–24].

There are also concerns about the costs of setting up,

running and maintaining LAF theatres as opposed to non-

LAF theatres [25].

However, in a recent review, Evans [26] suggests that

LAF has been accepted as an international standard for

arthroplasty [16] and that it is difficult to conclusively

deduce the role of LAF when the infection rates in joint

replacement surgery are multi-factorial. He points out that

LAF in the absence of overall operating theatre etiquette

like improper positioning of patients, personnel and

instruments under the LAF, not wearing surgical masks,

excessive operating room traffic with frequent opening and

closing of theatre doors intra-operatively tends to cause a

paradoxical increase in infection rates [27–30]. He further

argues that the cost of construction and installation of a

LAF theatre system has fallen over the years [31]. He feels

that this can now be justified as the cost of a single total hip

revision that averages 4.8 times higher than the direct

medical costs associated with primary total hip arthroplasty

[32, 33]. Consequently, the overall savings due to fewer

infections would ultimately outweigh the cost implications

of LAF theatres.

Table 2 Summary of patient demographics for hip arthroplasty in the two hospitals

Characteristics Hospital A (LAF) Hospital B (non-LAF) p value

Age 0.009

Range 19–100 years 25–96 years

Median 70 years 71 years

Sex 0.93

Male 38.5% 38.4%

Female 61.5% 61.6%

ASA score \0.001 I 14.6% 12.6%

II 66.3% 59.6%

III 18.5% 27.1%

IV 0.5% 0.6%

V 0.1% 0.06%

ASA American Society of Anaesthesiologists

Table 3 Summary of patient demographics for knee arthroplasty in the two hospitals

Characteristics Hospital A (LAF) Hospital B (non-LAF) p value

Age 0.46

Range 25–94 years 23–94 years

Median 70 years 70 years

Sex 0.07

Male 44.0% 41.5%

Female 56.0% 58.5%

ASA score \0.001 I 11.5% 11.0%

II 70.6% 62.8%

III 17.5% 25.3%

IV 0.23% 0.6%

V 0.1% 0.4%

ASA American Society of Anaesthesiologists

Eur J Orthop Surg Traumatol (2017) 27:261–265 263

123

Previous studies have shown that national joint registries

are invaluable for investigating rare complications such as

prosthetic joint infection as they record large number of

patients [10, 34–36]. These data would be extremely dif-

ficult to produce in prospective randomised trials. Fur-

thermore, joint registries and SSI databases record

procedures as they happen in real time in the clinical sitting

and so provide a more accurate picture of the patient than

under controlled conditions [13].

Our study takes its data from the NJR and our local

SSI database. The NJR data have been captured from

April 2003 to October 2014 and now represent more

than 11 years’ worth of hip and knee replacements

carried out at Hospitals A and B. These hospitals serve

a similar patient demographic and catchment area. The

skillset of surgeons working at these hospitals is the

same, and the hospitals share the same trust protocols

for infection control and theatre staff training. As a

result, factors such as patient mix, surgical technique,

use of pre- and post-operative antibiotics, movement

within the theatre and general sterile procedures would

be similar.

Our study shows that there was no statistically signifi-

cant difference between the two hospitals in terms of NJR

data for revision rates. Similarly, there was no significant

difference for our locally collected SSI data. These results

are surprising as there have been various studies that have

shown that the use of LAF reduces the number of colony-

forming units per cubic metre [27]. In turn, this should

reduce the infection rate. However, there have been no

studies to date that have demonstrated conclusively that a

reduced number of colony-forming units relates to a sim-

ilarly lower rate of infection [10].

As far as we are aware, this is the first study that

compares NJR and SSI data for LAF and non-LAF theatres

in hospitals serving a similar population with staff having

similar surgical skills and following the same sterilisation

and protocols.

Our results corroborate the findings of other recently

published papers [10, 13, 18, 20] and question the rationale

of increasing use of LAF in routine lower limb arthroplasty.

We agree with Uçkay et al. [37] in believing that pre-

vention of SSI in arthroplasty has unique hallmarks not

shared with other surgical specialities. Chief among them

are low inoculum for implant infections, difficulty in

eradicating infection in the presence of prosthesis, low

bone penetration of current antibiotics and pathogenicity of

coagulase-negative staphylococci. Only some of the many

measures to prevent orthopaedic SSI are based on strong

evidence.

The main limitations of our study are that it is between

only two hospitals and although the patient populations are

largely similar, it does not take into account the risk

adjustment and analysis of co-variants that might influence

the outcome of interest.

We possibly cannot draw policy recommendations from

this study but it is a pragmatic study that highlights an

urgent need for greater debate and more robust studies on

the advantages and disadvantages of LAF theatres in rou-

tine arthroplasty cases.

Acknowledgements We thank the patients and staff of all the hospitals in England, Wales and Northern Ireland who have contributed data to

the National Joint Registry. We are grateful to the Healthcare Quality

Improvement Partnership (HQIP), the NJR Research Sub-committee

and staff at the NJR Centre for facilitating this work. We also thank Mr

Paul Bassett and Ms Tara Nikovskis for their help with the statistics and

with editing the manuscript, respectively. We thank the Editor of the

Annals of the Royal College of Surgeons for giving us the permission to

use our SSI data previously published in this journal. The authors have

conformed to the NJR’s standard protocol for data access and publi-

cation. The views expressed represent those of the authors and do not

necessarily reflect those of the National Joint Registry Steering Com-

mittee or the Health Quality Improvement Partnership (HQIP) who do

not vouch for how the information is presented.

Compliance with ethical standards

Conflict of interest The authors declare that they have no conflicts of interest.

Ethical standards This article does not contain any studies with human participants or animals performed by any of the authors.

References

1. Biring GS, Kostamo T, Garbuz DS et al (2009) Two-stage revi-

sion arthroplasty of the hip for infection using an interim artic-

ulated Prostalac hip spacer: a 10- to 15- year follow-up study.

J Bone Joint Surg (Br) 91-B:1431–1437

2. Walenkamp G (2009) Joint prosthetic infections: a success story

or a continuous concern? Acta Orthop 80:629–632

3. Bannister G (2002) Prevention of infection in joint replacement.

Curr Orthop 16:426–433

4. Dreghorn CR, Hamblen DL (1989) Revision arthroplasty: a high

price to pay. BMJ 296:648–649

5. Merollini KMD, Crawford RW, Whitehouse SL (2013) Surgical

site infection prevention following total hip arthroplasty in

Australia: a cost-effective analysis. Am J Infec Control

41:803–809

6. Charnley J (1972) Postoperative infection after total hip

replacement with special reference to air contamination in the

operating room. Clin Orthop Related Res 87:167–187

7. Charnley J (1964) A clean-air operating enclosure. Br J Surg

51:202–205

8. Nelson JP (1976) Five years’ experience with operating room

clean rooms and personnel-isolator systems. Med Instrum

10:277–281

9. Marotte JH, Lord GA, Blanchard JP et al (1987) Infection rate in

total hip arthroplasty as a function of air cleanliness and antibiotic

prophylaxis. 10-year experience with 2384 cementless Lord

Madreporic prostheses. J Arthroplast 2:77–82

10. Hooper G, Rothwell A, Frampton C et al (2011) Does the use of

laminar flow and space suits reduce early deep infection after

total hip and knee replacement? The ten-tear results of the New

Zealand joint registry. J Bone Joint Surg (Br) 93-B:85–90

264 Eur J Orthop Surg Traumatol (2017) 27:261–265

123

11. Lidwell O, Lowbury E, Whyte W et al (1982) Effect of ultraclean

air in operating rooms on deep sepsis in the joint after total hip or

knee replacement: a randomised study. Br Med J 285:10–14

12. Miner A, Losina E, Katz J et al (2007) Deep infection after total

knee replacement: impact of laminar airflow systems and body

exhaust suits in the modern operating room. Infec Control Hosp

Epidemiol 28:222–226

13. Gastmeier P, Breier AC, Brandt C (2012) Influence of laminar

airflow on prosthetic joint infections: a systematic review. J Hosp

Infec 81:73–78

14. Merollini KMD, Zheng H, Graves N (2013) Most relevant

strategies for preventing surgical site infection after total hip

arthroplasty: guideline recommendations and expert opinion. Am

J Infec Control 41:221–226

15. Namba RS, Inacio MCS, Paxton EW (2012) Risk factors asso-

ciated with surgical site infection in 30,491 primary total hip

replacements. J Bone Joint Surg (Br) 94-B:1330–1338

16. Proceedings of the international consensus meeting on peripros-

thetic joint infection. (2013): Available at https://www.efort.org/

wp-content/uploads/2013/10/Philadelphia_Consensus.pdf. Acces-

sed Sept 29 2015

17. Humphreys H (2012) Surgical site infection, ultraclean ventilated

operating theatres and prosthetic joint surgery: where now?

J Hosp Infect 81:71–72

18. Brandt C, Hott U, Sohr D et al (2008) Operating room ventilation

with laminar airflow shows no protective effect on the surgical

site infection rate in orthopedic and abdominal surgery. Ann Surg

248:695–700

19. Assadian O, Kuelpmann R, Zhumadilova A et al (2009) Protec-

tive effect of HEPA-filtered operating room air ventilation with

or without laminar airflow on surgical site infections. Ann Surg

250:659–660

20. Breier AC, Brandt C, Sohr D et al (2011) Laminar airflow ceiling

size: no impact on infection rates following hip and knee pros-

thesis. Infec Control Hosp Epidemiol 32:1097–1102

21. Salvati EA, Robinson RP, Zeno SM et al (1982) Infection rates

after 3175 total hip and total knee replacements performed with

and without a horizontal unidirectional filtered air-flow system.

J Bone Joint Surg (Am) 64:525–535

22. Kurz A, Sessler DI, Lenhardt R (1996) Perioperative normoth-

ermia to reduce the incidence of surgical-wound infection and

shorten hospitalization. N Engl J Med 334:1209–1215

23. Melling A, Ali B, Scott E et al (2001) Effects of preoperative

warming on the incidence of wound infection after clean surgery:

a randomised controlled trial. Lancet 358:876–880

24. Forbes S, Eskicioglu C, Nathens A et al (2009) Evidence-based

guidelines for prevention of perioperative hypothermia. J Am

Coll Surg 209:492–503

25. Cacciari P, Giannoni R, Marcelli E et al (2004) Cost evaluation of

a ventilation system for operating theatre: an ultraclean design

versus a conventional one. Ann Ig 16:803–809

26. Evans RP (2011) Current concepts for clean air and total joint

arthroplasty: laminar airflow and ultraviolet radiation. A sys-

tematic review. Clin Ortop Relat Res 469:945–953

27. Andersson AE, Petzold M, Bergh I et al (2014) Comparison

between mixed and laminar airflow systems in operating rooms

and the influence of human factors: experiences from a Swedish

orthopedic center. Am J Infec Control 42:665–669

28. Panahi P, Stroh M, Casper DS et al (2012) Operating room traffic

is a major concern during total joint arthroplasty. Clin Orthop Rel

Res 470:2690–2694

29. Dickson GH (2001) Deterioration of theatre discipline during

total joint replacement: have theatre protocols been abandoned?

Ann R Coll Surg Eng 83:296

30. Taylor GJ, Bannister GC (1993) Infection and interposition

between ultraclean air source and wound. J Bone Joint Surg (Br)

75:503–504

31. Haslam KR (1974) Laminar air-flow air conditioning in the

operating room: a review. Anesth Analg 53:194–199

32. Bozic KJ, Kurtz SM, Lau E et al (2010) The epidemiology of

revision total knee arthroplasty in the United States. Clin Orthop

Related Res 468:45–51

33. Bozic KJ, Ries MD (2005) The impact of infection after total hip

arthroplasty on hospital and surgeon resource utilization. J Bone

Joint Surg (Am) 87:1746–1751

34. Willis-Owen CA, Konyves A, Martin DK (2010) Factors

affecting the incidence of infection in hip and knee replacement.

An analysis of 5277 cases. J Bone Joint Surg (Br) 92:1128–1133

35. Hooper GJ, Hooper NM, Rothwell AG et al (2009) Bilateral total

joint arthroplasty: early results from the New Zealand National

Joint Registry. J Arthroplasty 24:1174–1177

36. Rothwell AG, Hooper GJ, Hobbs A et al (2010) An analysis of

the Oxford hip and knee scores and their relationship to early

joint revision in the New Zealand Joint Registry. J Bone Joint

Surg (Br) 92-B:413–418

37. Uçkay I, Hoffmeyer P, Lew D et al (2013) Prevention of surgical

site infections in orthopaedic surgery and bone trauma: state-of-

the-art update. J Hosp Infec 84:5–12

Eur J Orthop Surg Traumatol (2017) 27:261–265 265

123

European Journal of Orthopaedic Surgery & Traumatology is a copyright of Springer, 2017. All Rights Reserved.