Screening for methicillin-resistant Staphylococcus Aureus (MRSA) carriage is a common practice in elective orthopaedic surgery. Preoperative decolonization protocols have proven to be effective in reducing MRSA carrier rates.1,2 Multiple large prospective studies in elective orthopaedic surgery have shown a significant decrease in the number of MRSA surgical site infections (SSIs) with implementation of MRSA screening and decolonization protocols.3,4 In addition to decolonization, appropriate antibiotic selection and dosing is an important factor in preventing MRSA-related SSI. Among MRSA-colonized patients, those receiving inadequate preoperative antibiotic coverage have significantly higher SSI rates.5,6
The overwhelming majority of the published orthopaedic literature investigating MRSA screening and decolonization protocols has been focused on elective orthopaedic patient populations such as arthroplasty, sports medicine, and spine surgery.7 In the setting of elective orthopaedic surgery, MRSA screening often occurs at a preoperative clinic visit, allowing sufficient time for identification of MRSA carriers and implementation of decolonization protocols before surgery. Acute orthopaedic trauma presents a unique scenario in which injuries may require urgent surgery that occurs outside of normal daytime hours.8 In these scenarios, MRSA colonization status is often unknown at the time of surgery.6,9 Even if carrier status were to be known at the time of surgery, a 5-day surgical delay to allow for decolonization is impractical in most orthopaedic trauma scenarios.
Previous studies have documented rates of MRSA colonization between 1.8% and 3.8% on orthopaedic trauma wards.6,10,11 The purpose of the current study was to prospectively determine the MRSA carrier rate among surgically treated orthopaedic trauma patients. We also aimed to determine the rates of SSI (overall and MRSA-related) in this patient population. In addition, we sought to determine differences in carrier rates and infection rates between acute orthopaedic trauma patients (those admitted through the emergency department) and patients undergoing elective surgery for sequelae of orthopaedic trauma.
PATIENTS AND METHODS
During the study period, the orthopaedic trauma service at the University of Wisconsin Hospital consisted of 4 fellowship-trained orthopaedic trauma surgeons. This service provides surgery for inpatient trauma, outpatient fracture care, and elective orthopaedic trauma-related conditions including deformity correction, nonunion, and malunion surgery. Over a 3-month period between April 15, 2016 and July 15, 2016, a new MRSA screening protocol was initiated for all patients undergoing surgery by any of the orthopaedic trauma surgeons. Nasal swabs (BBL CultureSwab; Becton, Dickinson and Company, Franklin Lakes, NJ) were used to obtain samples from patients' bilateral nares. Samples were tested for the presence of MRSA using a polymerase chain reaction (PCR) assay (GeneXpert; Cepheid, Sunnyvale, CA). If possible, patients were screened in the outpatient orthopaedic trauma clinic preoperatively when undergoing nonurgent surgery. Patients who presented to the emergency department with acute orthopaedic trauma injuries were screened on hospital admission. For patients who screened negative for MRSA colonization or those for whom test results were unavailable at the time of surgery, a weight-based dose of a first- or second-generation cephalosporin [cefazolin 2 g (3 g if ≥120 kg) or cefuroxime 1.5 g (3 g if ≥120 kg)] was administered every 8 hours for 24 hours, perioperatively. For all patients who screened positive for MRSA colonization, a weight-based dose of vancomycin was also administered preoperatively (20 mg/kg, rounded to the nearest 250 mg, with a maximum dose of 2 g), and institutional isolation protocols were followed.
Patient demographics, medical comorbidities, tobacco use, and substance abuse were recorded. Standard postoperative follow-up was conducted per the treating surgeon's protocol. All patients with less than 4 weeks of postoperative follow-up were included in the analysis of screening and colonization rates, but were excluded from any additional analysis related to infection. Several patients underwent multiple procedures during the screening period, with one patient undergoing 3 procedures. These patients were counted only once, and as such, determination of each patient's carrier status was counted as a single screening event. Infection was defined broadly as any superficial wound complication requiring oral antibiotics or any return to the operating room with positive bacterial cultures.
For statistical analysis, patients were divided into 2 main groups. The first group consisted of patients who underwent surgical management of an acute fracture in either the inpatient or outpatient setting. This group was labeled the “acute orthopaedic trauma” cohort. The second group, labeled “nonacute,” consisted of patients undergoing nonunion or malunion correction, heterotopic ossification excision, or other revision surgeries including removal of painful implants. The “nonacute” group also included patients with suspected infection who presented with draining or erythematous wounds near previous orthopaedic trauma implants and those undergoing eradication surgery for known chronic osteomyelitis or other infectious conditions. For statistical analysis, the “nonacute” group was further divided into 2 subgroups—“infected” patients (those undergoing infection-eradication procedures for documented or suspected infection) and “elective” patients (those undergoing surgical procedures for sequelae of orthopaedic trauma in whom infection was not documented or suspected). A flowchart for classification of patients into the various subgroups is shown in Supplemental Digital Content 1 (see Figure, http://links.lww.com/JOT/A838).
Patient variables were summarized by group using frequency (%), mean (SD), or median (interquartile range) as appropriate for each variable's distribution. Comparisons were made between the acute orthopaedic trauma and nonacute groups and between the acute orthopaedic trauma and elective groups using χ2 tests, t-tests, or Wilcoxon rank sum tests as appropriate for each variable's distribution. Logistic regression models were used to estimate the odds ratio between groups for screening positive for MRSA, and Cox proportional hazards models were used to estimate the hazard ratios for time to development of any infection between groups. Analyses were performed using R for Statistical Computing, version 3.5.1 (R Foundation for Statistical Computing, Vienna, Austria) and all tests were conducted at a 0.05 significance level.
Overall MRSA Carrier Rate
Over the 3-month study period, 248 individual patients underwent surgery by one of the 4 orthopaedic trauma surgeons. MRSA screening was performed on 71% (175/248) of these patients (Fig. 1). Six patients were found to be MRSA positive, resulting in an overall MRSA carrier rate of 3.4% (6/175).
Of the 248 individuals who had surgery during the study period, 82% (203/248) underwent surgical management of an acute fracture. The remaining 18% (45/248) of patients were treated for nonacute orthopaedic trauma conditions (Fig. 2). Within this “nonacute” group of 45 patients, 9 underwent surgical treatment of a documented or suspected infection. The remaining 36 patients constituted the “elective” group that underwent an orthopaedic trauma procedure not for acute fracture or infection. Surgical procedures in this group included removal of painful implants, heterotopic ossification excision, and surgical procedures to address nonunion or malunion (see Table, Supplemental Digital Content 2, http://links.lww.com/JOT/A839).
Overall, 27 of the 248 patients (11%) who underwent surgery during the study period had less than 4 weeks of follow-up and were excluded from further analysis, leaving 89% of patients (221/248) with minimum follow-up of 4 weeks. After removal of the 9 patients treated for known infection, 212 patients remained for the infection analysis, and all were included regardless of their MRSA screening status. Of these 212 patients with adequate follow-up, 180 were in the acute orthopaedic trauma group and 32 were in the elective group. As shown in Table 1, there were no differences between the acute orthopaedic trauma, nonacute, and elective groups with regard to sex, medical comorbidities, substance abuse, or length of follow-up. The average age of the acute orthopaedic trauma group was significantly older than the other groups with an average age of 55 compared to 46 in the nonacute and elective groups (P = 0.006 and 0.018, respectively, Table 1).
MRSA Carrier Rate by Cohort
In the acute orthopaedic trauma cohort, 143/203 patients (70%) were screened for MRSA. This group had an MRSA nasal carriage rate of 1.4% (2/143). In the nonacute group, 71% (32/45) were screened for MRSA. This cohort had a significantly higher MRSA carrier rate of 12.5% (4/32, P = 0.01) compared with the acute group. In the elective group, 72% (26/36) patients were screened, and similarly, the MRSA carrier rate was significantly higher compared with that of the acute group (15.4% vs. 1.4%, P = 0.005, Fig. 3). Compared with acute orthopaedic trauma patients, individuals in the elective group had 12.8 times increased odds (95% confidence interval, 2.36–96.5) of testing positive for MRSA.
Overall Infection Rate
During the study period, 9 individuals underwent surgical treatment for eradication of previous infections (chronic osteomyelitis in 3 patients and draining wounds suspicious for infection in 6 patients). After excluding these 9 patients with preexisting infection, the overall infection rate during the study period for all patients undergoing orthopaedic trauma procedures who had a minimum follow-up of 4 weeks was 2.4% (5/212). Two of these infections occurred in the acute orthopaedic trauma cohort, for an overall infection rate of 1.1% (2/180). The other 3 infections occurred in the elective cohort, resulting in a significantly higher overall infection rate of 9.4% (3/32), P = 0.025.
MRSA Infection Rate
The overall MRSA infection rate was 1.4% (3/212) (Fig. 3). Of patients with adequate follow-up, only one MRSA infection was identified in the acute orthopaedic trauma group (1/180, 0.6%), occurring in a patient who tested negative for MRSA colonization. The elective cohort had a higher MRSA infection rate of 6.3% (2/32), but this difference did not reach statistical significance (P = 0.29, Fig. 3).
Overall, 6/175 patients were colonized with MRSA (Table 2). Of these 6 patients, the MRSA status was unknown at the time of surgery for 2 patients (33%)—one in the acute orthopaedic trauma group and one in the elective group. Because the MRSA status was not known preoperatively, these patients received inadequate antibiotic prophylaxis (ie, no MRSA coverage). One of these 2 patients in the elective group went on to develop a postoperative MRSA infection. The other patient in the acute orthopaedic trauma group did not develop an SSI postoperatively. Of the 4 patients whose MRSA positive status was known preoperatively, all 4 received adequate preoperative antibiotics with vancomycin and cefuroxime. Despite this, one of these 4 patients in the elective group went on to develop a postoperative MRSA infection. Of the 143 patients in the acute orthopaedic trauma group who were screened, only one patient received inadequate antibiotic prophylaxis, and this patient did not develop a postoperative infection.
MRSA Colonization Rates
Several large studies of elective orthopaedic surgical patients have reported rates of MRSA colonization between 2.9% and 4.4%.2–4 In the orthopaedic trauma literature, one large study of 2473 patients admitted to an orthopaedic trauma unit identified a carrier rate of 3.2%.10 A separate study of 1122 trauma admissions reported an MRSA colonization rate of 3.8%.11 Our overall MRSA colonization (3.4%) is in keeping with the rates reported in the literature.
Despite best efforts to screen, MRSA colonization status may not be known at the time of fracture surgery, particularly in patients with hip fractures, which have reported MRSA carrier rates ranging from 1.8% to 16.9%,9,12–14 In a single center European study, adherence to best practice guidelines for early hip fracture surgery resulted in 86% of hip fractures reaching the operating room with unknown MRSA status.9 In a separate series of 400 orthopaedic trauma patients, only 41% of patients had their MRSA results before surgery despite testing on admission, and 81% of the MRSA + patients did not receive adequate preoperative antibiotic coverage.6 Among patients not receiving adequate MRSA coverage, 22.7% (5/22) developed an MRSA SSI compared with no infections in patients receiving adequate MRSA coverage (0/5).6
Although preoperative decolonization protocols may be important, appropriate perioperative antibiotic administration seems to be equally crucial. According to our institutional policy, adequate preoperative antibiotic coverage for MRSA + patients includes weight-based vancomycin dosing in addition to a first- or second-generation cephalosporin. There is some evidence that vancomycin administration in addition to another antibiotic such as cefazolin or clindamycin may not reduce SSIs or MRSA-specific SSIs following surgery; however, this may be due to insufficient dosing.15,16 Specifically, unless a larger loading dose is used, vancomycin's steady state is not reached until the fourth dose on an every-12-hour dosing regimen.17 It is possible, therefore, that common preoperative antibiotic regimens recommended for MRSA + patients are ineffective, and this particular question represents an area for future study.
Acute Orthopaedic Trauma
The acute nature of orthopaedic trauma presents unique challenges for MRSA screening protocols, as expeditious surgery may be necessary for urgent fracture stabilization or to reduce fracture-associated morbidity and mortality.8 However, the MRSA colonization rate among acute orthopaedic trauma patients at our institution was quite low (2/143, 1.4%). Although our results do not definitively discount the potential efficacy of MRSA screening in acute orthopaedic trauma patients, among populations with a low colonization rate, the number of patients receiving inadequate antibiotic prophylaxis because of unknown preoperative MRSA status is quite low.
Based on high infection rates in patients with inappropriate antibiotics, Iqbal et al6 advocated for routine antibiotic coverage for MRSA in all orthopaedic trauma patients with unknown MRSA status. Other authors have advocated for preoperative nasal povidone-iodine for decolonization purposes, and Tonne et al demonstrated a decreased overall infection rate with universal preoperative povidone-iodine nasal decolonization in an orthopaedic trauma population.18,19 Based on the study by Iqbal et al,6 use of prophylactic antibiotics for MRSA coverage would be needed in 227 patients to prevent one MRSA infection. Our results do not support routine vancomycin dosing of patients with unknown MRSA status, as only 0.7% of screened acute orthopaedic trauma patients (1/143) in our study received inadequate preoperative antibiotic prophylaxis. There were no infections in the acute cohort that could be attributed to inadequate MRSA antibiotic coverage. However, future research may delineate patient subgroups that could benefit from routine MRSA coverage if colonization status is unknown at time of surgery.
Operative treatment of acute fractures often results in the need for secondary procedures related to the initial treatment. These secondary surgeries may be performed to promote healing, correct deformity, remove implants, or eradicate infection. Among patients undergoing elective orthopaedic trauma surgery without a documented or suspected infection, the rate of MRSA colonization (15.4%) was significantly higher than previously reported in “elective” orthopaedic patient populations (2.9%–4.4%).2–4 This may be due to the increased health care exposure of this patient population. Of the 4 MRSA + patients in the elective group, 50% had multiple hospitalizations within the previous year. Consequently, this patient population may benefit from a screening and decolonization protocol to help reduce SSI caused by MRSA.
MRSA infection in the setting of orthopaedic trauma is associated with significant morbidity including extended hospital stays, multiple procedures, and high costs.11 PCR screening allows for rapid detection of MRSA colonization with relatively high sensitivity.20–22 However, if results are unobtainable before surgery, the utility of the test is lost. The current practice at our institution allows MRSA PCR testing to be run in real time during daytime hours from 7 AM to 3 PM, with results available after 2–3 hours. However, during evenings and weekends, result times vary widely. It is likely that many institutions face similar challenges related to decreased laboratory staffing and equipment availability during off hours. If laboratory protocols were optimized to facilitate rapid analysis and reporting of MRSA screening swabs at all times of the day, it may be possible for all but the most emergent trauma patients to have their MRSA status known before surgery.
Some authors advocate for universal hospital MRSA screening, whereas others advocate for targeted screening based on known MRSA risk factors.23 Documented risk factors in the literature include previous hospitalization within 24 months, residence at a long-term care or rehabilitation facility within past 18 months, surgical intervention within the past 60 months, antibiotic use within past 12 months, previous MRSA colonization, and the presence of skin wounds, ulcers, or indwelling catheters.23,24 According to previous studies, hip fracture patients are at high risk for MRSA colonization, with reported rates of up to 17%.12–14 In our study, however, only 2.9% of acute hip fractures (1/34) were MRSA positive. Further research is necessary to determine whether a selective screening policy based on patient risk factors is beneficial in the orthopaedic trauma patient population.
Some orthopaedic traumatic injuries may be treated on a delayed outpatient basis or treated with early temporizing stabilization, followed by delayed definitive surgical fixation. This delayed treatment of acute injuries may allow additional time for screening, decolonization, and appropriate selection of antibiotic prophylaxis. Given the success in the literature of screening and decolonization in other orthopaedic patient groups, it is possible that patients with nonurgent fractures could benefit similarly from decolonization.
As shown in Figure 1, 29% of patients (73/248) were not screened. Lack of screening occurred most frequently during the initial phase of the protocol implementation, mainly due to unfamiliarity with the screening protocol. Although our rates of MRSA carriage are in line with previous studies, the relatively small size of our study (175 patients screened) and the relatively brief duration of the screening period (3 months) limits our ability to definitively assess the true value of MRSA screening for this population. Our study is also limited by the short average duration of follow-up (4.1 months, range 2–15). However, we reviewed all patients' medical records a minimum of 12 months after the last patient was screened to capture any complications that occurred within the first year after screening. Further research is necessary to study the cost effectiveness of an MRSA screening protocol in larger cohorts of orthopaedic trauma patients. In addition, MRSA carriage rates vary among institutions and geographic regions, thus affecting the utility and efficacy of screening in different locations.
A second limitation of our study is related to the methods of testing for MRSA colonization. The screening protocol in our study consisted only of nasal swabs sent for PCR assays. Isolated nasal swabs have been shown to have moderate sensitivity, varying between 62% and 89%.17,18 Newer assays have reportedly higher sensitivities, up to 91%.19 There is also evidence that pooling of swabs from different body parts including the groin, axilla, and throat can increase sensitivity and detection rates.18 As such, it is possible that false negative screening results were present in our study, in which case the true MRSA carrier rate in our patient population would be higher than we have reported.
Finally, the “elective” group included patients with nonunion, a population that has been shown to harbor occult infections by nonculturable organisms in a biofilm phase of growth. If any of the patients with nonunion that developed a postoperative infection had a pre-existing, but unrecognized infection, this could theoretically result in misclassification bias, and such patients should have been excluded from the “elective” group. As such, the observed difference in overall infection rate between the elective group and the acute orthopaedic trauma group (9.4% vs. 1.1%, P = 0.025) may actually be smaller.
In this study, acute orthopaedic trauma patients have a low incidence of MRSA colonization and a low likelihood of receiving inadequate preoperative antibiotic prophylaxis because of unknown MRSA colonization status. The rate of MRSA colonization is significantly higher in patients undergoing nonacute orthopaedic trauma procedures. This patient group may benefit from an MRSA screening and decolonization protocol before surgery in addition to administration of appropriately selected antibiotics. However, our results do not support routine vancomycin dosing for patients with an unknown preoperative MRSA status, given the improbability of patients receiving inadequate antibiotic prophylaxis. Further research is needed to determine the most effective and cost-efficient way to reduce MRSA infection in the orthopaedic trauma patient population.
1. Chen AF, Heyl AE, Xu PZ, et al. Preoperative decolonization
effective at reducing staphylococcal colonization in total joint arthroplasty patients. J Arthroplasty. 2013;28:18–20.
2. Moroski NM, Woolwine S, Schwarzkopf R. Is preoperative staphylococcal decolonization
efficient in total joint arthroplasty. J Arthroplasty. 2015;30:444–446.
3. Sporer SM, Rogers T, Abella L. Methicillin-resistant and methicillin-sensitive Staphylococcus aureus screening
to reduce surgical site infection in elective total joint arthroplasty. J Arthroplasty. 2016;31:144–147.
4. Kim DH, Spencer M, Davidson SM, et al. Institutional prescreening for detection and eradication of methicillin-resistant Staphylococcus aureus in patients undergoing elective orthopaedic surgery. J Bone Joint Surg Am. 2010;92:1820–1826.
5. Murphy E, Spencer SJ, Young D, et al. MRSA
colonisation and subsequent risk of infection despite effective eradication in orthopaedic elective surgery. J Bone Joint Surg Br. 2011;93:548–551.
6. Iqbal HJ, Ponniah N, Long S, et al. Review of MRSA screening
and antibiotics prophylaxis in orthopaedic trauma
patients; the risk of surgical site infection with inadequate antibiotic prophylaxis in patients colonized with MRSA
. Injury. 2017;48:1382–1387.
7. Chen AF, Chivukula S, Jacobs LJ, et al. What is the prevalence of MRSA
colonization in elective spine cases? Clin Orthop Relat Res. 2012;470:2684–2689.
8. Sheehan KJ, Sobolev B, Guy P. Mortality by timing of hip fracture
surgery: factors and relationships at play. J Bone Joint Surg Am. 2017;99:e106.
9. Bryson DJ, Gulihar A, Aujla RS, et al. The hip fracture
best practice tariff: early surgery and the implications for MRSA screening
and antibiotic prophylaxis. Eur J Orthop Surg Traumatol. 2015;25:123–127.
10. Shukla S, Nixon M, Acharya M, et al. Incidence of MRSA
surgical-site infection in MRSA
carriers in an orthopaedic trauma
unit. J Bone Joint Surg Br. 2009;91:225–228.
11. Nixon M, Jackson B, Varghese P, et al. Methicillin-resistant Staphylococcus aureus on orthopaedic wards: incidence, spread, mortality, cost and control. J Bone Joint Surg Br. 2006;88:812–817.
12. Merrer J, Pisica-Donose G, Leneveu M, et al. Prevalence of methicillin-resistant Staphylococcus aureus nasal carriage among patients with femoral neck fractures: implication for antibiotic prophylaxis. Infect Control Hosp Epidemiol. 2004;25:515–517.
13. Thyagarajan D, Sunderamoorthy D, Haridas S, et al. MRSA
colonisation in patients admitted with hip fracture
: implications for prevention of surgical site infection. Acta Orthop Belg. 2009;75:252–257.
14. Gessmann J, Kammler J, Schildhauer TA, et al. MRSA
colonisation in patients with proximal femur fractures in a German trauma centre: incidence, infection rates and outcomes. Langenbecks Arch Surg. 2012;397:117–123.
15. Saveli CC, Morgan SJ, Belknap RW, et al. Prophylactic antibiotics in open fractures: a pilot randomized clinical safety study. J Orthop Trauma. 2013;27:552–557.
16. Ponce B, Raines BT, Reed RD, et al. Surgical site infection after arthroplasty: comparative effectiveness of prophylactic antibiotics: do surgical care improvement project guidelines need to be updated?. J Bone Joint Surg Am. 2014;96:970–977.
17. Alexander JW, Solomkin JS, Edwards MJ. Updated recommendations for control of surgical site infections. Ann Surg. 2011;253:1082–1093.
18. Torres EG, Lindmair-Snell JM, Langan JW, et al. Is preoperative nasal povidone-iodine as efficient and cost-effective as standard methicillin-resistant Staphylococcus aureus screening
protocol in total joint arthroplasty? J Arthroplasty. 2016;31:215–218.
19. Tonne B, Humphrey A, Smith S, et al. Preoperative nasal povidone-iodine solution effectively reduces the rate of surgical site infection in orthopaedic trauma
cases. Podium Presentation, Annual Meeting—Orthopaedic Trauma
Association (2018), Orlando, FL.
20. Senn L, Basset P, Nahimana I, et al. Which anatomical sites should be sampled for screening
of methicillin-resistant Staphylococcus aureus carriage by culture or by rapid PCR test? Clin Microbiol Infect. 2012;18:E31–E33.
21. Blanc DS, Nahimana I, Zanetti G, et al. MRSA screening
by the Xpert MRSA
PCR assay: pooling samples of the nose, throat, and groin increases the sensitivity of detection without increasing the laboratory costs. Eur J Clin Microbiol Infect Dis. 2013;32:565–568.
22. Yarbrough ML, Warren DK, Allen K, et al. Multicenter evaluation of the xpert MRSA
NxG assay for detection of methicillin-resistant Staphylococcus aureus in nasal swabs. J Clin Microbiol. 2018;56:e01381–e01317.
23. Coia JE, Leanord AT, Reilly J. Screening
for meticillin resistant Staphylococcus aureus (MRSA
): who, when, and how? BMJ. 2014;348:g1697.
24. Xue Y, Gyi AA. Predictive risk factors for methicillin-resistant Staphylococcus aureus (MRSA
) colonisation among adults in acute care settings: a systematic review. JBI Libr Syst Rev. 2012;10:3487–3560.