Audit of Positive Microbiological Cultures in Patients Undergoing Arterial Reconstruction
Article Outline
Objective
To describe the micro-organisms cultured in vascular patients and report the changing spectrum of pathogens over recent years.
Methods
Retrospective analysis of post-operative microbiology culture results from wound swabs, blood cultures, theatre specimens and central venous catheter tips for all patients undergoing vascular procedures between 1998 and 2003.
Results
460 patients underwent a major vascular procedure; suprainguinal bypass (n
=34), infrainguinal bypass (n=158), carotid endarterectomy (n=97) and abdominal aortic aneurysm repair (n=171). Positive microbiology results were identified in 86 patients (18%). Patients admitted as emergency had a significantly higher (chi2=20.2, p<0.0001) proportion of positive cultures (29% vs. 13%).The proportion of positive cultures was significantly different between procedures; suprainguinal bypass 29%, infrainguinal bypass 27%, abdominal aortic aneurysm repair 19%, carotid endarterectomy 0% (chi2=33.9, p<0.0001). The most commonly isolated organisms were Methicilin-sensitive Staphylococcus aureus (26%), MRSA (19%), and Staphylococcus epidermidis (10%) accounting for infection in 78% of patients. The prevalence of positive cultures declined over the study period from 18% in 1998 (18%) to 9% in 2003. No cases of MRSA were identified before 1999, compared to 12 cases in 2000–2001 (8.5%) and 11 cases in 2002–2003 (7.1%). There was no significant relationship between positive culture status and in-hospital mortality.
Conclusion
Staphylococci were most common after vascular procedures and MRSA was cultured in a significant proportion. Antimicrobial chemoprophylaxis should aim to cover these organisms.
Keywords: Positive microbiological cultures, Staphylococcus aureus, Arterial reconstruction
Introduction
Graft infection is an infrequent complication of vascular surgery. Overall infection rates are in the region of 5%, but this incidence varies according to site of the graft.1, 2 Infection rates are higher in emergency and revision surgery, and when a groin anastomosis has been constructed. A graft infection can have devastating consequences for the patient and is associated with considerable morbidity, including the risk of limb loss. Mortality rates are also high and figures of 40–75% have been quoted for aortic bypass graft and 10% for femoro-popliteal graft infection.
Staphylococcus Aureus is the most common organism to infect a graft in the post-operative period. These highly virulent infections tend to occur within four months of surgery however in the last fifteen years the incidence of low virulence micro-organisms that appear many months to years post-op has increased. Staphylococcus epidermidis and gram negative bacteria (Escherichia coli, Pseudomonas, Klebsiella) are found in 40% and mixed gram-positive and gram-negative occur in 10–15% of cases.3
Patients with critical ischaemia, gangrene or ulceration are likely to have colonisation or frank infection of the skin with pathogens capable of infecting the wound. Recent evidence from the Joint Vascular Research Group in 2001 showed that Methicillin-resistant Staphylococcus aureus (MRSA) is the leading cause of vascular infections in the UK and was found in 40% of patients with wound infections and 33% of graft infections.4 Patients with MRSA positive wound infections were significantly more likely to die or undergo amputation than MRSA negative infections.
Randomised double blind trials demonstrated that antibacterial prophylaxis reduced the incidence of wound infection following reconstructive arterial surgery.5, 6 A first or second generation cephlasporin, or co-amoxiclav, is the most common choice for prophylaxis in arterial surgery, with aminoglycosides providing an alternative in those that are allergic to beta-lactam antibiotics.
We aim to study only those types of colonisation which could result in a graft infection. Therefore we only considered surgical site infections, blood cultures or positive cultures of indwelling central venous catheters. This study aims to identify the most common pathogens isolated in patients undergoing major vascular procedures and to study whether there has been a change in these micro-organisms in the last five years.
Methods
Data collection
All patients undergoing major arterial reconstruction at our centre over a five-year period between January 1998 and December 2003 were entered into the study with a minimum follow-up of 18 months. Post-operative positive cultures in these patients were audited. No cultures were taken at the primary procedure. The indications for post operative cultures were clinical suspicion of infection i.e. wound erythema, wound ooze, pyrexia, rising white cell count and exit site inflammation of central venous catheters. These patients were identified using the trust's local computer based coding system, providing information on date of admission, date of surgery, procedure and date of discharge. Patient identification numbers were matched to laboratory records in order to obtain microbiology culture and sensitivity (C+S) results from operative specimens, post-operative wound swabs, blood cultures and central venous catheters within the immediate six-month period following surgery. Hospital notes were reviewed to provide information regarding follow-up and outcome.
Statistical analysis
Statistical analysis was done with The Stata 8.1 for windows. Significance was set at p<0.05.
Results
A total of 460 patients underwent major arterial reconstruction within the study period. The mean age of the population was 71 years (range 30 to 91). Eighteen percent (n=84) of the study population had positive microbiology cultures. There was no significant difference in the proportion of positive results in patients above or below median age of the population. The annual positive culture rate decreased significantly over the study period (Table 1), (test for trend, p=0.02).
Table 1. Culture results and number of major vascular procedures each year
| Year | Number of procedures | Negative cultures | Positve cultures |
|---|---|---|---|
| 1998 | 74 | 56 (76%) | 18 (24%) |
| 1999 | 91 | 73 (80%) | 18 (20%) |
| 2000 | 74 | 58 (78%) | 16 (22%) |
| 2001 | 67 | 56 (84%) | 11 (16%) |
| 2002 | 89 | 74 (83%) | 15 (17%) |
| 2003 | 65 | 59 (91%) | 6 (9%) |
| Total | 460 | 376 (82%) | 84 (18%) |
The procedures performed were 34 supra-inguinal bypass, 158 infra-inguinal bypass, 97 carotid endarterectomy (CEA) and 171 abdominal aortic aneurysm repair (AAA). Table 2 shows the distribution of positive and negative cultures between patients undergoing different procedures. No positive cultures were found in patients undergoing carotid endarterectomy.
Table 2. Distribution of positive and negative cultures between different procedures
| Procedure | Number of procedures | Negative culutures | Positive cultures |
|---|---|---|---|
| Supra inguinal bypass | 34 | 24 (71%) | 10 (29%) |
| AAA repair | 171 | 138 (81%) | 33 (19%) |
| CEA | 97 | 97 (100%) | 0 (0%) |
| Infrainguinal bypass | 158 | 117 (74%) | 41 (26%) |
| Total | 460 | 376 | 84 |
The percentage of patients with positive microbiology was significantly higher in those patients who were admitted as an emergency: 47 out of 160 (29%) in patients who underwent an emergency admission and 37 out of 300 patients (12%) who had elective surgery. (chi2=20.3, p<0.001).
Methicillin-sensitive Staphylococcus aureus (MSSA), Methicillin-resistant Staphylococcus aureus (MRSA) and Staphyloccus epidermidis were the most common organisms identified on culture and represented 66% of all positive results (Table 3). No cases of MRSA were identified before 2000. The annual proportion of patients with positive cultures for MRSA was 0% in 1998 and 1999, 8% in 2000, 9% in 2001, 9% in 2002 and 5% in 2003 (chi2=15.4, p=0.009). MRSA was significantly more common in emergency admissions (11% vs. 2%, chi2=20.2, p<0.001). Seventy eight percent of patients with a MRSA positive culture were admitted as an emergency. The median length of stay was significantly greater in patients with a positive culture: 18 days (IQR: 13–34) compared to those with no positive cultures: 9 days (IQR: 5–15) (p<0.001, Wilcoxon ranksum test). There was no significant difference in length of hospital stay between patients who had a positive culture for MRSA and those who had a positive culture for other micro-organisms.
Table 3. The most common organisms and their frequencies
| Organism | Number of cultures | % of total |
|---|---|---|
| Methicillin-sensistive Staphylococcus aureus | 32 | 38 |
| Methicillin-resistant Staphlococcus aureus | 23 | 27 |
| Staph. epidermidis | 12 | 14 |
| Streptococci | 6 | 7 |
| Escherichia coli | 5 | 6 |
| Other | 6 | 7 |
| Total | 84 | 100 |
Some 49% of positive cultures were obtained in the first 10 days following surgery and 85% were obtained by the 30th post-operative day. There was no significant relationship between positive culture status and in-hospital mortality. Overall, 2 patients had graft infections, which needed removal of the graft. One patient had an MRSA infected AAA graft and a second patient had a femoral-popliteal graft removed following wound dehiscence. No growth was obtained from the graft.
Discussion
A positive culture from a surgical significant site, such as blood cultures, central venous catheters or surgical site infection was found in almost one fifth of vascular patients. The majority of these infections were caused by staphylococci. MRSA had been cultured in a significant proportion of patients and seemed to be increasing. Fortunately, a prosthetic graft infection remained an uncommon complication. Our AAA graft infection rate was 0.6% which compares accurately to published series (0.5–1%). Our infra-inguinal graft infection rate 0.6% which is lower than published series (3–6%).1, 2 This may reflect more stringent antibiotic prophylaxis or a lower emergency case load. It is recognised that the incidence of infection is much greater in emergency surgery as compared with elective surgery and our study confirms this. Factors such as extended peri-operative hospital stay predisposing to virulent pathogens, less stringent adherence to perioperative prophylaxis or sterile technique, increased incidence of bowel injury, more extensive lymphatic manipulation or injury during dissection, repeat operations and prolonged operating times could all contribute to this.7, 8, 9
Our study shows that Staphylococci are responsible for the vast majority of positive cultures (MSSA 38%, MRSA 27%, Staphylococcus epidermidis 14%). The remaining cultures were mainly Streptococcus and Escherichia coli. Our results compare similarly with the reported literature.3 Therefore optimal chemo-prophylaxis should cover these organisms. The prevalence of MRSA appears to be increasing as is shown in our results. This increase is reflected in results reported from other vascular units. Our results do show a decline in the positive culture rate between 2002 and 2003 – this is not reflected in the literature but may be the result of increased use of anti-MRSA prophylaxis in our unit.
No study has demonstrated clear advantage of glycopeptide prophylaxis or prophylaxis with newer drugs with activity against MRSA such as linezolid over standard beta-lactam prophylaxis. Adequate trials assessing glycopeptide usage in prophylaxis for arterial surgery are lacking.10, 11, 12
It has previously been reported that the single most important determinant of MRSA infection is length of hospital stay.13 We have demonstrated in our series that the length of stay is significantly longer in those patients that developed positive cultures, but no difference was found between MRSA positive cultures and positive cultures with other micro-organisms. Almost 80% MRSA positive were admitted as an emergency. Our data suggest that preoperative screening for MRSA of surgical admissions should include emergency admissions. Opportunities to screen emergency admissions for MRSA with traditional methods are limited. However a new rapid detection of MRSA directly from nasal swab specimens has recently been described.14 This assay can give results within 2 hours and would allow more tailored infection control measures in emergency admissions. However adequate surgical prophylaxis against MRSA in patients who undergo emergency vascular operations remains important until its widespread introduction in the healthcare facilities.
From our study it is difficult to draw firm conclusions on which prophylactic antibiotics to use since our positive cultures may represent colonisation as well as infection. In our retrospective analysis of specimens a significant number of MRSA positive cultures were isolated from a wide range of sources including wound swabs. It is possible that these patients were colonised with MRSA prior to surgery and their presence in cultures post-operatively does not necessarily imply infection. However, pre-op carriers of MRSA are at significant risk of developing MRSA infection. The EMMS (Evaluation of molecular MRSA screening) study began in November 2005 and this involves screening all admissions to surgical wards for MRSA. This may give us a better idea of how many patients are colonised with MRSA prior to surgery and this can be used to guide an antibiotic prophylaxis regimen.
Conclusion
Our study shows the majority of surgically significant positive cultures following major vascular procedures are caused by staphylococci. MRSA has been cultured in a significant proportion of patients.
References
- . Antibiotic prophylaxis in clean surgery – vascular surgery. J Chemother. 2001;13(1):123–128
- . Management of patients with prosthetic vascular graft infection. Am Surg. 2000;66(2):166–177
- Vascular prosthetic graft infection. Epidemiology, bacteriology, pathogenesis and treatment. Acta Chir Belg. 2002;102:238–247
- . The impact of MRSA on vascular surgery. Eur J Vasc Endovasc Surg. 2001;22:211–214
- . Antibiotic prophylaxis in vascular surgery. Ann Surg. 1978;188:283–289
- . Effects of prophylactic antibiotics in vascular surgery. A prospective, randomized, double-blind study. Ann Surg. 1984 Jul;200(1):86–92
- . Management of infected aortic grafts. Surg Clin North Am. 1995;75(4):783–797
- In situ replacement of infected aortic grafts with rifampicin-bonded prostheses: the Leicester experience (1992 to 1998). J Vasc Surg. 1999;30(1):92–98
- . Heroic procedures in vascular injury management: the role of extra-anatomic bypasses. Surg Clin North Am. 2002;82(1):115–124
- Prospective randomised double blind trial comparing teicoplanin and cephazolin as antibiotic prophylaxis in prosthetic vascular surgery. Eur J Clin Microbiol Infect Dis. 1999;18:175–178
- . A comparison of Teicolplanin versus cephradine plus metronidazole in the prophylaxis of postoperative infection in vascular surgery. J Hosp Infect. 1999;41:233–244
- Antibacterial prophylaxis in surgery: 3–arterial surgery in the abdomen, pelvis and lower limbs. Drug Ther Bull. 2004;42(6):43–47
- The acquisition of methicillin-resistant Staphylococcus aureus (MRSA) in vascular patients. Eur J Vasc Endovasc Surg. 2003;25(2):147–151
- . Detection of methicillin-resistant Staphylococcus aureus directly from nasal swab specimens by a real-time PCR assay. J Clin Microbiol. 2004;42(12):5578–5581
PII: S1533-3167(06)00065-3
doi:10.1016/j.ejvsextra.2006.06.002
© 2006 Elsevier Ltd. All rights reserved.
