Is There a Time-Dependent Contamination Risk to Open Surgical Trays During Total Hip and Knee Arthroplasty?

Abstract

Background

Periprosthetic joint infection (PJI) after total hip and knee arthroplasty (TJA) is a devastating complication and intraoperative contamination can be a source for PJI. Currently, many measures are performed intraoperatively to reduce the risk of contamination. The primary purpose of this study was to determine if there is a time-dependent risk of contamination to open sterile surgical trays during TJA cases.

Methods

A prospective intraoperative culture swab study was performed. Standard sterile operating room trays without instruments were utilized as the experimental trays. These were opened simultaneously with all other surgical instrumentation needed for the procedure. These trays were left on an isolated Mayo stand next to the scrub tech’s table and swabbed at 30-minute intervals. The first swab was performed immediately after opening all sets and the last swab performed on closure of the incision. A new section of the grid-lined tray was swabbed for each data point and the culture analysis was conducted by our institutions’ microbiology lab for both quantitative and qualitative analysis. Operating suite room temperature and humidity data was also gathered.

Results

Twenty-three consecutive primary TJA cases in high air turnover rooms were included. 13 of the 23 (57%) cases demonstrated culture positive bacterial growth on at least one time point. Of the 109 independent swabs collected, 19 (17%) had bacterial growth. The most common bacterial species isolated was Staphylococcus epidermidis. There were no statistically significant associations between time (p= 0.35), operating room (OR) temperature (p = 0.99), and OR humidity (p = 0.07) and with bacterial growth.

Conclusion

In spite of isolating an organism in 57% of cases, we could not identify a time-dependent increase in bacterial contamination throughout our operative cases. We were unable to associate OR environmental temperature and humidity to bacterial growth.

Level of Evidence: II

Introduction

Infection following total joint arthroplasty is a devastating complication. Contamination intraoperatively is thought to be a main cause of acute periprosthetic joint infections (PJI). PJI typically necessitates subsequent surgery, and has high morbidity to the patient.¹ This poses a very significant burden on both the patient and the treating surgeon to minimize infection risk.

Arthroplasty surgeons use protocols to mitigate contamination on the day of surgery including: performing operations in high air turnover rooms, clipping skin day of operation, meticulously prepping skin with both alcohol and non-alcohol preparations, carefully draping patients in a specific fashion, using occlusive iodine impregnated dressing to cover skin before incision, wearing hooded gowns, changing gloves immediately after draping, avoiding OR traffic, and operating efficiently so as to minimize incision time.² While intuitively these steps should lower potential contamination, little data have been published with regards to these steps reducing postoperative infection. Further, infection rates have remained relatively constant despite these steps.² There have been many studies that have investigated intraoperative risk factors associated with PJI. Time of procedure has been shown by multiple studies to be associated with the development of PJI.^3,4 However, these studies do not specify where the source of the infection originated from. It is reasonable to assume that longer operative times lead to more opportunities for contamination of the sterile field, but there is limited data on if surgical trays are the source of contamination leading to PJI.

We sought to identify additional potential interventions that may decrease the infectious risk to patients. One variable that was identified was selective opening of operating room trays based upon when they are needed in the procedure. Previous studies have shown that there is a time dependent rate of contamination of trays intraoperatively, but this has been specific to spine fusion procedures.^5-7 Little to no data has been published specific to joint arthroplasty, a unique surgical procedure in which deep infection has particularly devastating consequences including potential need for a two-stage revision arthroplasty and long-term IV antibiotics. It is uncertain as to if contamination of trays occurs in arthroplasty rooms, rooms that already take precautions to reduce air contamination and potential settling of debris on open surgical trays. The purpose of this study was to swab and culture this potential contamination site and determine if delaying the opening of trays until they are necessary can help to mitigate bacterial contamination loads. Our hypothesis is that a temporal relationship between bacterial contamination of surgical trays and case time will be identified.

Methods

Control Set Preparation and Swabbing

After obtaining institutional review board exemption status, we selected one senior total joint arthroplasty surgeon and obtained culture results from his operating room. Twenty-three consecutive primary total joint arthroplasty cases were included in this analysis. Standard sterile operating room trays without instruments were utilized. These were opened simultaneously with all other surgical instrumentation needed for the procedure. These trays were left on an isolated Mayo stand next to the scrub nurse’s table and swabbed at 30-minute increments by a research assistant not actively scrubbed into the case. The first swab was performed immediately after opening all sets and the last swab performed on closure of the incision. A new section of the grid-lined tray was swabbed for each data point and the culture analysis was conducted by our institutions’ microbiology lab for both quantitative and qualitative analysis. Operating suite room temperature and humidity data was also gathered.

Microbiological Analysis

For the bacterial analysis portion, we utilized a modified protocol described by Shams et al.⁸ The samples were collected using the BD ESwab collection kit (Becton Dickenson) The swab was removed and moistened with sterile saline. The tray samples were collected by applying gentle but firm pressure using back and forth vertical S-strokes with one side of the swab and back and forth horizontal S-strokes with the other side of the swab, using templates to standardize surface sampling. After placing the swabs into collection tubes, breaking off the shafts and replacing the caps, the swabs were delivered to the laboratory, refrigerated, and processed within 24 hours. The tubes were vortexed for 5 seconds.

The samples were spread 100μL each of a 1:10 and 1:100 dilution in Tryptic soy broth (TSB) per University of Iowa micro lab standard operating protocol in both a non-selective and selective agar using an L-shaped cell spreader. The agar plates were incubated at 35° for 48 hours, and the colonies counted and recorded on plates with between 20 and 250 colonies. The number of colony forming units (CFUs) per sample, based on the dilutions performed were recorded. The isolated organisms were identified to species level using the MALDI-TOF mass spectrometry (Bruker). The isolates were banked at -80 degrees and subsequently destroyed after 48 hours of culture analysis.

Statistical Analysis

Rates of contamination was reported with descriptive statistics. This was then analyzed from a statistical perspective with Mann-Whitney U test and Wilcoxon Sign Ranks tests. Room temperature and humidity were separately analyzed to evaluate for significance. Categorical variables were evaluated with chi-square test or Fisher’s exact test. A p-value < 0.05 was considered statistically significant.

Results

Twenty-three consecutive total joint arthroplasty cases from one orthopaedic surgeon were analyzed with swabs sent for culture and quantitative analysis. In addition to culture results, additional data points including room temperature and humidity were also analyzed. The complete data sets are included in Table 1. The numerical values represent the number of colony forming units observed under microscopic analysis. In some of these analyses, there were two separate bacterial colonies that were noted, and these are noted by the duplicate measurements. Additionally, the bacterial colonies that speciated out on culture analysis are also included. For some of the cases, there are only four data points that are included implying that the case had been completed in less than 120 minutes.

Table 1.

Outlines Each Surgical Case With the Species of Bacteria Found

Case #	T1 (CFU)	T2 (CFU)	T3 (CFU)	T4 (CFU)	T5 (CFU)
1	*NGTD	NGTD	1000 /10 (M. luteus)	NGTD	10
2	NGTD	NGTD	NGTD	100 (M. oleivorans) /10 (S. capitis)	NGTD
3	NGTD	NGTD	NGTD	NGTD	NGTD
4	NGTD	100 (S. acidaminiphila) /20	**NQ (S hominis)	NGTD	NGTD
5	NGTD	NGTD	NGTD	NGTD	10 (C. tuberculosis)
6	NGTD	NQ (S. epidermidis)	NGTD	NGTD	NGTD
7	NGTD	NGTD	NGTD	NGTD
8	NGTD	NGTD	NQ	NGTD	NGTD
9	NGTD	NGTD	NGTD	NGTD	NGTD
10	NGTD	NGTD	NGTD	NGTD	NGTD
11	NGTD	NGTD	NGTD	NGTD	NGTD
12	NGTD	NQ (M. testaceum)	NGTD	NGTD	NGTD
13	NGTD	NGTD	NQ (S. epi)	10/10 (C. jeikeium)	NGTD
14	NGTD	10 (S. capitis)	NQ (S. hominis)	NQ (S. epi)	NGTD
15	NGTD	NGTD	NGTD	NGTD	NGTD
16	NGTD	NGTD	NGTD	NGTD	NGTD
17	NGTD	NGTD	NGTD	NQ (M. luteus)
18	NGTD	NGTD	NGTD	NGTD
19	NGTD	NGTD	NGTD	NGTD	NGTD
20	NGTD	NGTD	NGTD	NGTD
21	NGTD	NQ (S epidermidis)	NGTD	NGTD
22	30 (M. luteus)	NGTD	NGTD	NGTD	NGTD
23	10 (S epi- dermidis) /10 (S. hominis)	NQ (S. epidermidis)	NGTD	NGTD

*No Growth to Date (Ngtd)

**Nq Represents That Bacteria Was Noted on the Culture

Analysis, But These Were Unable to be Quantified.

Of these 23 cases, 13 of the cases (57%) demonstrated culture positive bacterial growth on at least one time point. The time-points in which each positive culture was obtained and detected is outlined in Figure 1, with the highest number of positive cultures being at 30 minutes. Of the 109 independent swabs collected, 19 (17%) had bacterial growth. The most common bacterial species isolated was Staphylococcus epidermidis. OR temperatures varied from 62 to 69 F and humidity varied from 38 to 55% relative humidity. There were no statistically significant associations between time (p= 0.35), OR temperature (p = 0.99), and OR humidity (p = 0.07) and bacterial growth.

Figure 1.

Frequency of positive cultures as a function of time.

As outlined in Figure 2, our research team found that the most commonly isolated species in our study was S. epidermidis but this was only present in 6 cultures out of 109. The next most common bacteria isolated was Staphylococcus hominis and Micrococcus luteus and was present in 3 of the 109 cultures. Other bacteria isolated were Microbacterium oleivorans, Stenotrophonomonas acidaminiphila, Corynebacterium tuberculosteatricum, Microbacterium testaeum, and Corynebacterium jeikeium each of which were only isolated in 1 culture of the 109. As for the quantitative number of bacteria grown, the most CFU found were 1000, and the least was 10.

Figure 2.

Frequency of isolated bacterial contaminants.

Discussion

The purpose of this study was to identify a temporal relationship of case length and OR tray contamination. Our hypothesis was that longer case length would be associated with higher frequency of OR tray contamination. Although, there was contamination of OR trays in 57% of cases, it did not have a temporal relationship. This went against our original hypothesis and previous studies that have shown time dependent increases in bacterial contamination as a basis of time spent in the operating rooms. Despite variances in operating room temperature and relative humidity, there was no association with presence of bacterial growth or bacterial load. Additional data is needed to determine if a temporal relationship exists for OR tray contamination in total joint arthroplasty cases.

Dalstrom et al. did reveal the time dependent nature of tray contamination in a controlled experimental setting and did show that covering the trays with a surgical towel was an effective way to prevent contamination. With other studies displaying the association with increased surgical time and infection rates we were surprised to find no association between time and contamination.^9,10 Although there is an abundance of data that has found the association between increased operating time and the prevalence of PJI, there is the need for investigation for further interventions to lower the rate of PJI.

There have been studies that have not shown a time-dependent risk for surgical site infections, specifically for total hip arthroplasty.¹¹ This suggests that there may be a difference in the risk of longer procedures and the risk of contamination for total knee arthroplasty and total hip arthroplasty. Regardless, it is known bacterial contamination of sterile surfaces occurs during total knee and hip arthroplasty, and there have been multiple interventions that have improved contamination rates. It is still yet to be discovered the if there is a relationship between surgical tray contamination and the rate of PJI.

Further investigation is warranted for the time-relationship between length of surgical case and contamination of sterile OR trays. Our data suggest that opening all surgical trays at the beginning of the case does not have a temporal relationship with contamination and remains to be a safe practice for total knee and hip arthroplasty.

Limitations

This study was cut down from the original cohort size of 100 patients based upon a lack of a time dependent relationship between time in OR and bacterial contamination in preliminary data analysis during our preliminary data analysis. If this had been carried out to completion, it is possible that a time dependent relationship could have been observed. An additional possible weakness is the fact that cultures were limited to 48 hours to decrease chance of contaminants. This was based upon our institutions’ microbiology policy and also based upon similar studies conducted previously.¹² There is always a chance that some species might have been isolated if culture were kept for longer time periods.

Conclusion

In summary, this study did not find a time-dependent relationship with contamination or OR trays during a TJA. Additionally, operating room temperature and humidity did not correlate with detection of tray contamination. The bacteria that were isolated also are known to be common causes of PJI. Investigating the relationship between surgical tray contamination and rate of PJI remains necessary. Also, additional data is warranted regarding staggered tray opening and rate of contamination during total knee and hip arthroplasty to ensure a change in practice is not warranted.