University of Groningen. Improving diagnostic accuracy in aortic prosthetic graft infection Saleem, Rani

Transcription

1 University of Groningen Improving diagnostic accuracy in aortic prosthetic graft infection Saleem, Rani IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2017 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Saleem, R. (2017). Improving diagnostic accuracy in aortic prosthetic graft infection [Groningen]: Rijksuniversiteit Groningen Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date:

2 Chapter 2 Conservative treatment of vascular prosthetic graft infection is associated with high mortality Am J Surg. 2010;200:47-52 Ben R. Saleem Robbert Meerwaldt Ignace F.J. Tielliu Eric L.G. Verhoeven Jan J.A.M. van den Dungen Clark J. Zeebregts

3 CHAPTER 2 Abstract BACKGROUND: The aim of this study was to identify patient-related and/or disease-related factors that influence outcomes in patients with vascular prosthetic graft infections. METHODS: Through the hospital patient administration system, between January 1997 and December 2007, a total of 44 patients were diagnosed with central prosthetic graft infections. Univariate and multivariate analyses were performed to define factors predictive of mortality. RESULTS: Thirty-three men and 11 women (mean age, 71 years) were included. There was considerable comorbidity. Coagulase-negative Staphylococcus and S. aureus were isolated in almost 50% of the patients. The mean follow-up duration was 5 years, during which 20 patients (46%) died. The main causes of death were related to vascular disease. Conservative treatment with antibiotics was the only variable with significant predictive value on multivariate analysis (hazard ratio, 3.62; 95% confidence interval, ; P =.02). CONCLUSIONS: Conservative treatment of prosthetic graft infections was associated with high mortality; therefore, it should be limited to a specific group. Patients who are not capable of undergoing open repair may benefit from conservative management. Otherwise, aggressive open treatment seems indicated. 20

4 CONSERVATIVE TREATMENT AND HIGH MORTALITY Introduction Prosthetic graft infection after vascular reconstruction is a rare but devastating and potentially life-threatening complication. 1 Despite the increasing success obtained with endovascular devices, Dacron â (DuPont, Wilmington, DE) and polytetrafluoroethylene are still being used on a routine basis for the open repair of both aneurysmal and occlusive atherosclerotic disease. Several measures can be taken to prevent the development of prosthetic graft infections, such as the use of perioperative prophylactic antibiotics, the avoidance of groin incisions, adherence to sterility, and the use of synthetic prosthetic grafts that are either rifampicin bonded or silver coated. Nevertheless, prosthetic grafts remain sensitive to infection. The incidence of graft infection has been reported to range from.6% to 3%, with a mortality rate varying from 25% to 88% and an amputation rate ranging from 5% to 25%. 1-4 Although several studies have been published, mainly reporting small series of prosthetic infections, there are no clear guidelines with regard to diagnostic and treatment management. Traditionally, the treatment of prosthetic graft infection has included antibiotics, excision of the infected prosthesis, and extra-anatomic bypass whenever appropriate. 5-7 This retrospective study was conducted to analyze clinical characteristics, diagnostic approaches, treatment strategies, and outcomes of this dreadful complication. The primary aim of the study was to determine if there are any factors predictive of mortality. Methods Between January 1997 and December 2007, a total of 56 patients with histories of vascular reconstruction were admitted for prosthetic graft infections to our tertiary referral center. Peripheral prosthetic graft infections (3 femorofemoral and 7 femoropopliteal grafts) and dialysis arteriovenous shunts were excluded from this study. Also, 2 patients with endograft infections were excluded from further analysis. A total of 44 patients with central prosthetic graft infections were the subjects of this retrospective study. Patients comorbidities were defined as recommended according to the Ad Hoc Committee on Reporting Standards. 8 Renal disease was defined as a serum creatinine level of 2.5 to 5.9 mg/dl, a creatinine level > 6.0 mg/dl, or on dialysis or with kidney transplant. 21

5 CHAPTER 2 The diagnosis of graft infection was based on the following criteria: (1) clinical evidence of infection (e.g. fever, pain, swelling, and elevated infection parameters); (2) evidence of graft infection on computed tomography (CT), magnetic resonance imaging, leukocyte scan, or 18 F-fluorodeoxyglucose positron emission tomography (PET) combined with CT; (3) operative findings (e.g. necrosis, purulent fluid, infected graft material); and (4) the isolation of microorganisms from blood, drain material, or the prosthesis. To gather the required information, the hospital s patient administration system was consulted. Also, general practitioners and patients were contacted for the completion of all required data. Primary vascular reconstructions were categorized as central or peripheral reconstructions; thoracoabdominal, aortoiliac, aortofemoral, and iliofemoral were considered central reconstructions and both femorofemoral and femoropopliteal as peripheral reconstructions. The underlying disease was categorized as aneurysmal or occlusive vascular disease. Time to clinical presentation of the infection was divided into early and late presentation. Early presentation related to presentation with one of the described criteria above <3 months after the primary operation, whereas late infection occurred after 3 months. Early mortality was defined as death <30 days after admittance for prosthetic infection or within the same hospital admission, and late mortality as death after 30 days. Statistical analyses were performed using SPSS for Windows version (SPSS, Inc, Chicago, IL). Data are presented as mean ± SD. Differences between categorical variables were tested using Pearson s c 2 test. Differences between means were tested using Student s 2-tailed test (for variables with normal distributions) or the Mann-Whitney U test (for those with skewed distributions). Statistical significance was set at P <.05. Univariate analysis with log-rank tests was performed to determine whether one or more patient-related and disease-related factors were predictive for mortality. Variables that had a direct influence after univariate analysis were entered in a stepwise backward manner into a multivariate Cox regression model. 22

6 CONSERVATIVE TREATMENT AND HIGH MORTALITY Table 1. Patient characteristics (n=44) Variable Value Sex Male 33 (75%) Female 11 (25%) Age (y) 71 (51-88) Body mass inde (kg/m 2 ) 26 (18-40) Comorbidities* Tobacco use 32 (73%) Hypertension 20 (46%) Cardiac disease 20 (46%) Hyperlipidemia 19 (43%) Renal disease 11 (25%) Pulmonary disease 11 (25%) Diabetes mellitus 9 (21%) Malignancy 7 (16%) Immune-suppressive therapy 6 (14%) Data are expressed as number (percentage) or as mean (range). *Definition according to the Ad Hoc Committee on Reporting Standards. 4 Results There were 33 men and 11 women (mean age, 71 years; age range, years). There was considerable comorbidity: almost half of the patients had cardiac disease, and one third had renal disease (Table 1). The revascularization procedures performed before the onset of infection are listed in Table 2. In 80% of cases, central reconstructions were involved. Seventy-five percent of the infected prostheses were initially implanted for aneurysmal disease (26% for ruptured aneurysms, 34% for acute nonruptured aneurysms, and 40% in the elective setting). Eight of the prostheses were initially implanted in primary infected aneurysms (i.e. mycotic aneurysms), and there was 1 inflammatory aneurysm. Ninety-eight percent of the patients with prosthetic graft infections presented with clinical symptoms. The most frequently encountered symptoms are listed in Table 3. In 86% of cases, causative microorganisms could be isolated, 50% of them cultivated from material retrieved after drain insertion and 48% from prosthetic graft material obtained during the operation (Table 4). The most 23

7 CHAPTER 2 common organisms cultured are listed in Table 3. Early clinical presentation, within 3 months, occurred in 43% of the patients. In most of these cases, coagulase-negative Staphylococcus or S. aureus was isolated. S. aureus and Escherichia coli tended to be more common in patients with late clinical presentation (mean infection-free interval, 33 months). The mean infection-free interval after the primary operation was 29 months (range, months). In our study, coagulase-negative Staphylococcus tended to be more common in early clinical presentation than other isolated microorganisms, with a mean infection-free interval of 21 months. The diagnosis was primarily made on the basis of a combination of the clinical presentation and additional imaging, especially CT (75%). In 30% of cases, CT was fused with 18 F-fluorodeoxyglucose PET. The remaining imaging modalities used to diagnose prosthetic graft infections are shown in Table 4. Various treatment modalities were used during the study period. The most applied treatment strategy was so-called conservative treatment, with the administration of antibiotics and close follow-up, followed by complete graft removal and in situ reconstruction. Details of other treatment strategies are described in Table 5. Table 2. Prosthesis-related factors at initial operation Factor n (%) Underlying disease Aneurysmal 33 (75) Occlusive 11 (25) Localization prosthesis Thoracoabdominal 1 (2) Aortoiliac 27 (61) Aortofemoral 10 (23) Iliofemoral 6 (14) Type of graft material Dacron â 41 (93) Polytetrafluoroethylene 3 (7) 24

8 CONSERVATIVE TREATMENT AND HIGH MORTALITY Table 3. Symptomatology and isolated microorganisms Variable n (%) Symptoms Pain 36 (82) Fever 22 (50) Swelling/suture aneurysm 17 (39) Wound infection 15 (34) Isolated microorganisms Coagulase-negative Staphylococcus 12 (27) S. aureus 8 (18) E. coli 5 (12) Candida albicans 3 (7) Pseudomonas aeruginosa 3 (7) Listeria monocytogenes 1 (2) Haemophilus influenzae 1 (2) Micrococcus spp 1 (2) Enterococcus faecalis 1 (2) Hafnia alvei 1 (2) No microorganism isolated 6 (14) Unknown 2 (5) The mean follow-up duration was 5 years, during which 20 patients (46%) died. Early mortality was seen in 4 patients (9%) after admittance for prosthetic graft infections. Three of them died because of multiple organ failure, and 1 patient died of massive bleeding from the aortic stump. Late mortality occurred in 16 patients (37%). The main cause of death in the late mortality period was also vascular disease related (n = 7 [16%]). Five patients died because of multiple organ failure. The other 2 patients died because of massive bleeding. Other causes of death in the late period were comorbidity related and were predominantly chronic obstructive pulmonary disease, cardiac disease, and malignant disease. The mortality reasons are listed in Table 6. Five-year prosthetic graft infection survival is shown in Figure 1. On univariate analysis with log-rank testing, from all patient-related or diseaserelated factors, conservative treatment was identified to be predictive of mortality. Factors that approached significance on univariate analysis were entered in a multivariate Cox regression model (backward wald). All other factors are shown in Table 7. On multivariate analysis, conservative treatment 25

9 CHAPTER 2 was the only significant predictive factor (hazard ratio, 3.62; 95% confidence interval, ; P =.02). Table 4. Diagnosis prosthetic graft infection Variable n (%) Clinical presentation 43 (98) Additional imaging 36 (82) CT 33 (75) PET 13 (30) Magnetic resonance imaging 2 (5) Leukocyte scan 5 (11) CT fused with PET 13 (30) Microorganism cultivated from Blood 9 (21) Drain production* 22 (50) Prosthesis 21 (48) *Production cultivated from the drain after the secondary operation Table 5. Treatment strategies Type n (%) Conservative treatment (antibiotics) 13 (30) Drainage 9 (20) Removal of graft and drainage 1 (2) Removal of graft and extra-anatomic bypass 8 (18) Removal of graft and in situ reconstruction Rifampicin drenched 13 (30) Table 6. Outcomes of treatment Variable n (%) Mortality 13 (30) Reason for overall mortality 9 (20) Vascular disaese related 1 (2) Pulmonary 8 (18) Renal 13 (30) Cardiac 2 (4) Other 5 (11) 26

10 CONSERVATIVE TREATMENT AND HIGH MORTALITY Table 7. Predictive factors for mortality (univariate analysis) Variable Log-rank P value Gender.81 Obesity.42 Diabetes mellitus.57 Smoking.62 Hypertension.48 Hypercholesterolemia.69 Cardiac disease.34 Renal disease.93 Pulmonary disease.31 Malignancy.84 Immune-suppressive therapy.88 Vascular disease.24 Extra-anatomic bypass.14 Primary groin anastomosis.76 Prosthetic material.18 Referring hospital.44 Antibiotics at admission.19 Fever.82 Pain*.45 Anastomotic aneurysm.88 Wound infection.49 S. aureus.54 CNS.40 Onset of symptoms after surgery.11 Conservative treatment.06 In situ reconstruction.76 CNS = Coagulase-negative Staphylococcus *After admission because of infection Factor approached significance 27

11 CHAPTER 2 Comments This retrospective analysis shows that the management of graft infection remains one of the most difficult challenges to vascular surgeons, mainly because of the heterogeneity of clinical presentation, comorbidities, and various treatment options available. Our results show that postoperative mortality in the studied group was substantial. Survival analysis showed 2-year cumulative survival of 80% and 5-year survival of 64% (Figure 1). These numbers correlate well with survival rates found in similar studies in the literature Figure 1. Kaplan-Meier cumulative survival curve describing 5-year follow-up of prosthetic graft infections. In our study, we tried to identify prognostic factors for mortality after admission for prosthetic graft infection. Conservative treatment was found to be predictive of mortality on univariate analysis. Prosthetic material, antibiotics at admission, the onset of symptoms after surgery, and extra-anatomic bypass were factors that approached significance on univariate analysis. The most important predictor of mortality on multivariate analysis was conservative treatment with antibiotics, indicating that this should not be an option for the management of 28

12 CONSERVATIVE TREATMENT AND HIGH MORTALITY prosthetic graft infection. There is a group of highly compromised patients, however, who could benefit from conservative treatment, such as those who cannot tolerate extensive surgical reconstruction or who have grafts in locations where they cannot be excised. 1 With regard to surgical treatment, several options exist. One is graft excision and extra-anatomic bypass, by placement of a new graft through an uninfected route, which is still widely accepted as the gold standard. In our study, this procedure was performed in 18% of patients, an approach that may be complicated by aortic stump blowout. 1-4 Graft excision and in situ replacement with either a rifampicin-bonded Dacron â prosthesis or an autogenous vein could be a good alternative. In particular, the use of an autogenous vein is claimed to be most effective in avoiding reinfection according to a meta-analysis comparing the clinical outcomes associated with 4 treatment modalities for aortic graft infection, including extra-anatomic bypass, rifampicin-bonded prosthesis, cryopreserved allograft, and autogenous vein. 2 Reconstruction with an autogenous vein is not a new technique. In 2002, Daenens et al. 13 published their 10-year experience with 49 patients showing the major advantage of the autogenous vein (ie, a low reinfection rate). 14 There are, however, disadvantages associated using an autogenous vein, including the need for preoperative duplex evaluation and the extended operative time needed to harvest these vessels, which may be less suitable for high-risk patients or in acute cases. 2,13 This operation is also contraindicated in those who have recanalized or profound dominant deep venous systems. 13 In almost 27% of the cases, coagulase-negative Staphylococcus was isolated (Table 3). It is known that a late presentation of prosthetic graft infection with coagulase-negative Staphylococcus can be managed with less extensive procedures. 5 However, in our patient population, prosthetic graft infections with coagulase-negative Staphylococcus tended to be more common in early clinical presentation compared with all other isolated microorganisms. There were no differences in the type of treatment used between prosthetic grafts infected with coagulase-negative Staphylococcus and other microorganisms. Diagnosing graft infections is not easy; symptoms are nonspecific, even though most patients are symptomatic. With sensitivity and specificity of 94% and 85%, respectively, CT is considered the best examination to confirm the diagnosis. 20 In our study, CT was not always sufficient to identify the exact localization of the source of infection. CT cannot distinguish perigraft abcedation from sterile perigraft fluid, and it has a high false-positive rate, especially within the first 6 29

13 CHAPTER 2 weeks after the operation To overcome this problem, we fused 18 F- fluorodeoxyglucose PET with CT in 30% of cases. The fused PET/CT not only localized the suggestive lesions in all cases but also was very sensitive in determining the extent of the infections (Figure. 2). This information can be very useful in selecting the right type of treatment. There were several limitations in this study. It was not a comparative study, and the number of patients was small, but prosthetic graft infection is an uncommon event. In view of the small number, the contributions of different risk factors to the etiology of graft infection is difficult to evaluate. This study included a heterogenic group; different graft materials and different treatments were used. Mainly because of the heterogeneity of the group, prosthetic graft infection management should be highly individualized. Because of the limitations of this study, we feel that more research with a larger volume of cases is warranted. Conclusions The diagnosis of vascular prosthesis graft infection remains difficult, mainly because of the heterogeneity of the group in its clinical presentation and the duration and type of antibiotics used, with decreased diagnostic sensitivity as a consequence. CT is the first choice for diagnosing prosthetic graft infection, although fused PET/CT has been proved and seems very promising for the near future. The treatment of a prosthetic graft infection should be highly individualized. There is no one best treatment for a prosthetic graft infection, but the use of autogenous vein reconstruction has gained popularity because of its low mortality and morbidity. Therefore, in an individual patient with a prosthetic vascular graft infection, many things must be considered to appropriately determine the treatment most likely to achieve eradication of the infection and long-term limb salvage with the lowest risk. Considering that conservative treatment is associated with high mortality, it should be used limited to a specific group. The establishment of a multicenter registry to record such complications is needed to confirm the findings of this study. 30

14 CONSERVATIVE TREATMENT AND HIGH MORTALITY Figure 2. (A) Computed tomographic image (transversal plane) of a 72-year-old man after an abdominal aortic aneurysm with an aortoiliac Dacron â prosthesis, which was infected with E. coli. The images illustrate an infection of the right leg of the prosthesis. (B) Same image as in (A) after fusion with 18 F-fluorodeoxyglucose positron emission tomographic image shows extension of infection to both legs, including the left groin. This patient was treated with extra-anatomic reconstruction after complete graft removal. 31

15 CHAPTER 2 References 1. Perera GB, Fujitani RM, Kubaska SM. Aortic graft infection: update on management and treatment options. Vasc Endovascular Surg 2006;40: O Connor S, Andrew P, Batt M, et al. A systematic review and meta-analysis of treatments for aortic graft infection. J Vasc Surg 2006;44: Swain TW III, Calligaro KD, Dougherty MD. Management of infected aortic prosthetic grafts. Vasc Endovascular Surg 2004;38: Seeger JM. Management of patients with prosthetic vascular graft infection. Am Surg 2000;66: Wilson SE. New alternatives in management of the infected vascular prosthesis. Surg Infect (Larchmt) 2001;2: Henke PK, Bergamini TM, Rose SM, et al. Current options in prosthetic vascular graft infection. Am Surg 1998;64: Zetrenne E, McIntosh BC, McRae MH, et al. Prosthetic vascular graft infection: a multi-center review of surgical management. Yale J Biol Med 2007;80: Rutherford RB, Baker JD, Ernst C, et al. Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg 1997;26: Lehnert T, Gruber HP, Maeder N, et al. Management of primary aortic graft infection by extraanatomic bypass reconstruction. Eur J Vasc Surg 1993;7: Gustavo S, Oderich MD, Thomas C, et al. Evolution from axillofemoral to in situ prosthetic reconstruction for the treatment of aortic graft infections at a single center. J Vasc Surg 2006;43: Zhou W, Lin P, Bush R, et al. In situ reconstruction with cryopreserved arterial allografts for management of mycotic aneurysms or aortic prosthetic graft infections: a multi-institutional experience. Tex Heart Inst J 2006;33: Gabriel M, Pukacki F, Checinski P, et al. Current options in prosthetic graft infection: comparative analysis of 63 consecutive cases. Lange- nbecks Arch Surg 2004;389: Daenens K, Fourneau I, Nevelsteen A. Ten-year experience in autogenous reconstruction with the femoral vein in the treatment of aortofemoral prosthetic infection. Cardiovasc Surg 2002;10: Nevelsteen A, Lacroix H, Suy R. Infrarenal aortic graft infection: in situ aortoiliofemoral reconstruction with the lower extremity deep veins. Eur J Vasc Endovasc Surg 1997;14(suppl):S88- S Vogt PR, Brunner-LaRocca HP, Lachat M, et al. Technical details with the use of cryopreserved arterial allografts for aortic infection: influence on early and midterm mortality. J Vasc Surg 2002;35: Verhelst R, Lacroix V, Vraux H, et al. Use of cryopreserved arterial homografts for management of infected prosthetic grafts: a multicentric study. Ann Vasc Surg 2000;14: Seeger JM, Pretus HA, Welborn MB, et al. Long-term outcome after treatment of aortic graft infection with staged extra-anatomic bypass grafting and aortic graft removal. J Vasc Surg 2000;32: Yeager RA, Taylor LM Jr, Moneta GL, et al. Improved results with conventional management of infrarenal aortic infection. J Vasc Surg 1999;30:

16 CONSERVATIVE TREATMENT AND HIGH MORTALITY 19. Chiesa R, Astore D, Piccolo G, et al. Fresh and cryopreserved arterial homografts in the treatment of prosthetic graft infections: experience of the Italian Collaborative Vascular Homograft Group. Ann Vasc Surg 1998;12: Orton DF, LeVeen RF, Saigh JA, et al. Aortic prosthetic infections: radiologic manifestations and implications for management. Radio- graphics 2000;20: Lawrence PF. Management of infected aortic grafts. Surg Clin North Am 1995;75: Vaart van der MG, Meerwaldt R, Slart RH, et al. Application of PET/SPECT imaging in vascular disease. Eur J Vasc Endovasc Surg 2008;35: Winter F, Vogelaers D, Gemmel F, et al. Promising role of 18-F-fluoro-Ddeoxyglucose positron emission tomography in clinical infectious diseases. Eur J Clin Microbiol Infect Dis 2002;21: Ducasse E, Calisti A, Speziale F, et al. Aortoiliac stent graft infection: current problems and management. Ann Vasc Surg 2004;18:

17