Long-Term Results with Cryopreserved Arterial Allografts (CPAs) in the Treatment of Graft or Primary Arterial Infections

Long-Term Results with Cryopreserved Arterial Allografts (CPAs) in the Treatment of Graft or Primary Arterial Infections

Journal of Surgical Research 168, e149–e153 (2011) doi:10.1016/j.jss.2010.09.026 Long-Term Results with Cryopreserved Arterial Allografts (CPAs) in t...

329KB Sizes 3 Downloads 55 Views

Journal of Surgical Research 168, e149–e153 (2011) doi:10.1016/j.jss.2010.09.026

Long-Term Results with Cryopreserved Arterial Allografts (CPAs) in the Treatment of Graft or Primary Arterial Infections Robert A. McCready, M.D.,1 M. Ann Bryant, M.S.N., R.N., A.C.N.P.–B.C., John W. Fehrenbacher, M.D., Daniel J. Beckman, M.D., Arthur C. Coffey, M.D., Joel S. Corvera, M.D., David A. Hormuth, M.D., and Thomas C. Wozniak, M.D. Department of Cardiovascular Surgery, Clarian Health Partners, Inc., Methodist Hospital, Indianapolis, Indiana Submitted for publication July 27, 2010

Objective. Our purpose was to evaluate our results with CPAs in patients with infected grafts or primary arterial infection with emphasis on long-term durability of these grafts. Methods. To evaluate the long-term durability of CPAs, clinical outcomes were analyzed following their use for either graft or primary arterial infections at a single institution over a 9-y period (2000–2009). The 30-d mortality rate, 90-d mortality rate, and the cause of early mortality were determined in each case. Among those surviving 90 d, the grafts were evaluated for subsequent failure. Results. From 2000 through 2009, 51 patients with either infected prosthetic grafts (35) or primary arterial infections (15) received CPAs. One patient had infection of a previously placed thoracic allograft. Forty-three graft infections involved either the thoracic or abdominal aorta. Eleven patients presented with fulminant sepsis with systemic inflammatory response syndrome (SIRS), seven of whom died postoperatively. Eight patients presented with aortoenteric, esophageal, or bronchial fistulae with infected prosthetic grafts. The 30-d mortality rate was 25.5% (11 deaths) seven of which occurred in patients with SIRS. The 90-d mortality rate was 41.4%. There were 10 graft failures, seven occurring in patients with aorto-enteric or bronchial fistulae all of whom had recurrent hemorrhage. The other three graft failures were due to anastomotic hemorrhage in the early postoperative period. Among those surviving 90 d, the mean follow-up was 46.4 mo (range 1–112 mo). No aneurysmal degeneration of the CPAs was noted. Only one

1

To whom correspondence and reprint requests should be addressed at Clarian Cardiovascular Surgeons, 1801 N. Senate Blvd. Ste. #755, Indianapolis, IN 46202. E-mail: [email protected].

subsequent allograft graft failure was noted among those surviving more than 90 d. Conclusions. CPAs are a suitable option in dealing with cardiovascular infections. Patients with enteric or bronchial fistulae are a difficult group to treat perhaps because of ongoing contamination of the allograft. The operative mortalities are largely determined by patient comorbidities (SIRS). Subsequent degeneration or infection of the CPAs is rare. Ó 2011 Elsevier Inc. All rights reserved. Key Words: cryopreserved allografts; graft infections; primary arterial infection; aorto-esophageal fistula; aorto-bronchial fistula; infected aneurysms.

INTRODUCTION

Primary arterial infections or infected prosthetic grafts are among the most challenging problems that cardiovascular surgeons must deal with. Prosthetic graft infections with involvement of the bronchus, esophagus, or gastrointestinal tract are especially difficult to treat in view of the potential for exsanguinating hemorrhage or ongoing contamination from the bronchial or intestinal defect. Patients with either primary arterial or prosthetic graft infections may present with an indolent infection or with profound sepsis which likely has a significant impact on the ultimate outcome. The concept of systemic inflammatory response syndrome (SIRS) and its relationship to outcome among patients with infected aneurysms suggests that the patients’ immune system plays an integral role in determining the ultimate outcome [1–3]. There are numerous reports describing the use of CPAs in treating these infections with generally

e149

0022-4804/$36.00 Ó 2011 Elsevier Inc. All rights reserved.

e150

JOURNAL OF SURGICAL RESEARCH: VOL. 168, NO. 1, JUNE 1, 2011

favorable results [4–14]. However, questions remain about long-term durability of CPAs and the potential for subsequent degeneration of the allografts. The primary aim of this study was to evaluate our experience in a single institution with CPAs in the treatment of infections involving the thoracic or abdominal aorta and the extra-coelomic arteries with particular emphasis on long-term results. PATIENTS AND METHODS From 2000 through 2009, patients treated at Methodist Hospital for either primary arterial infections or secondary infections involving prosthetic grafts, in whom CPAs were utilized, were reviewed and form the basis of this report. We treated 51 patients, 35 of whom had infected prosthetic grafts, 15 had primary arterial infection, and one patient had an infection of a previous thoracic allograft that had been placed at another institution. Among the 51 patients, the average age was 65.6 y (range 34–87). There were 35 males and 16 females. IRB approval was obtained for review of the medical records. The diagnosis of primary arterial infection or prosthetic graft infection was made on the basis of wound or blood cultures, radiographic, or operative findings. The location of the infections is listed in Table 1. All CPAs were obtained from CryoLife Inc., (Kennesaw, GA). No attempts were made for blood type matching. No immunosuppressive medications were used. Patients were maintained on intravenous antibiotics for 6 to 8 wk under the supervision of an infectious disease consultant. No patient was maintained on antibiotics indefinitely.

OPERATIVE DETAILS

Cultures of the artery and/or graft material were obtained in all patients. Aggressive removal of all infected or necrotic material was standard. Following arterial reconstruction, coverage of the allograft with omentum or muscle flaps was accomplished whenever feasible. Drains were placed selectively at the surgeon’s discretion. Surviving patients were encouraged to return for a CT scan on at least an annual basis. Additional follow-up was obtained from contact with the patients’ primary care physician, review of the social security death index, and telephone contact with the patient. MATERIALS AND METHODS Thoracic and Abdominal Aorta (43 Patients) Among these 43 patients, there were 31 patients with infected prosthetic grafts and one patient with an infected thoracic allograft that had been placed in another institution to treat an aorto-bronchial fistula. The patient was referred to us with recurrent hemorrhage and infection of the allograft. There were 11 patients with infected aneurysms involving the thoracic aorta (three patients) or abdominal aorta (eight patients). There were nine patients with either an aortobronchial fistula (two patients), aorto-esophageal (two patients), or aorto-enteric fistulae (five patients), all of whom presented with hemorrhage. All nine of these patients had infected prosthetic grafts. Eleven of these 43 patients presented with sepsis and met the criteria for SIRS.

TABLE 1 Anatomic Location of Infections 1. Thoracic aorta – (18 patients) a. Infected prosthetic grafts – 15 patients i. Aorto-esophageal fistulae with infected graft – 2 patients ii. Aorto-bronchial fistula with infected graft – 1 patient b. Infected allograft – 1 patient c. Primary infected thoracic aneurysm – 2 patients 2. Abdominal aortic infections – (26 patients) a. Infected prosthetic grafts – 18 patients i. Associated aorto-enteric fistulae – 5 patients b. Primary aortic infections – 8 patients i. Associated rupture – 5 patients 3. Peripheral arterial infections – (8 patients) a. Infected carotid-subclavian prosthetic graft – 1 patient b. Infected femoro-femoral prosthetic graft – 1 patient c. Infected axillo-femoral prosthetic graft – 1 patient d. Infected prosthetic femero-popliteal prosthetic grafts – 2 patients e. Infected femoral pseudoaneurysms – 3 patients

RESULTS

Among the 15 patients with thoracic aortic involvement, there were three deaths within 30 d, two of which resulted from anastomotic hemorrhage presumably secondary to residual infection. The third death was due to MSOF (multisystem organ failure). Two patients died within 90 d, neither of which was related to the allograft. There was one late allograft infection in a patient with an aorto-bronchial fistula who developed recurrent hemoptysis at 8 mo, which was treated by replacing the allograft with another allograft. Fifteen months later, recurrent hemoptysis developed. Further intervention was not undertaken because of multiple comorbidities, and the patient expired with recurrent hemorrhage. Among the 28 patient with abdominal aortic infections, there were eight deaths within 30 d and eight additional deaths within 90 d. Four of the five patients with aorto-enteric fistulae died secondary to recurrent hemorrhage. One additional graft failure occurred from anastomotic hemorrhage in the early postoperative period. All five allograft failures were due to anastomotic hemorrhage presumably secondary to residual infection. The remaining 11 deaths were due to either MSOF or overwhelming sepsis (SIRS). Among those surviving patients with abdominal allografts, no recurrent infections or allograft-related complications were noted (Fig. 1). One patient with end-stage renal disease on hemodialysis developed calcification of the allograft on long-term follow-up (Fig. 2). He died 90 mo postoperatively from a myocardial infarction. Extra-Celomic Infections (8 Patients)

Among these eight patients, one patient had an infected prosthetic carotid-subclavian graft, one1 patient

MCCREADY ET AL.: LONG-TERM RESULTS WITH CRYOPRESERVED ARTERIAL ALLOGRAFTS (CPAS)

e151

had an infected axillo-femoral graft, one1 patient had an infected femoro-femoral prosthetic graft, and two patients had infected prosthetic femoro-popliteal grafts. There were three infected femoral pseudoaneurysms, all secondary to angiographic procedures or cardiac catheterization with recurrent hemorrhage in two. There were no deaths and no graft failures among these eight patients.

DISCUSSION

There are many different treatment options when dealing with either infected prosthetic grafts or infected aneurysms. Traditional treatment has been vascular reconstruction through a non-contaminated route (extra-anatomic bypass) followed by removal of the infected graft or artery [13–15]. Remote bypass is not an option if the infection involves a critical anatomic area such as the aortic arch or visceral arteries. In these situations, in situ reconstruction is required. Aorto-enteric fistulae represent a difficult group of patients to treat regardless of the method of treatment. Four of our five patients with aorto-enteric fistulae involving the abdominal aorta died postoperatively, all of whom had recurrent hemorrhage. Recurrent aortoenteric fistulae were demonstrated in two of these patients at autopsy. One of our two patients with an aorto-esophageal fistula also died with recurrent bleeding. One of the two patients with an aortobronchial fistula developed recurrent hemoptysis 8 mo postoperatively and this patient represented the only late allograft failure. Other reports in literature indicate a high mortality rate with aorto-enteric fistulae. Six of the seven patients with aorto-enteric fistulae in the series reported by Nevelsteen et al., treated with CPAs died postoperatively [4]. In a multi-center study reported by Verhelst et al., involving CPAs, the mortality rate among the 17 patients with aorto-enteric fistulae was 83%, and eight of these 17 patients had postoperative disruption of the allografts presumably secondary to persistent infection [5]. Other authors have reported more favorable results using CPAs in treating aorto-enteric fistulae. Leseche et al. had seven survivors among 10 patients treated for aorto-enteric fistulae [9]. In their series of 49 patients with CPAs, Vogt et al. reported a 14% 30-day mortality rate for patients with aorto-enteric fistulae [15]. In dealing with these fistulae, Vogt and his colleagues recommended peritoneal drainage because of the severity of the infection. In addition, these authors circumferentially reinforced the anastomosis with allograft strips and covered the anastomosis with gentamicin-impregnated glue.

FIG. 1. CT scan demonstrating abscess formation around an infected prosthetic graft following repair of a ruptured abdominal aortic aneurysm in November 2008.

Although our mortality rates for patients with aorto-enteric fistulae, aorto-esophageal fistula, or aorto-bronchial fistulae were quite high, we believe that utilizing CPAs in these cases is reasonable. Meticulous closure of the intestinal or bronchial defect, adequate drainage, coverage of the allograft with omentum or a muscle flap, and provision of adequate nutritional support would seem to maximize the chances of a favorable outcome in this extremely difficult group of patients. A patient’s immune status likely plays a significant role in determining the ultimate outcome in a patient

FIG. 2. CT scan nearly 8 y later demonstrating minimal calcification of the allograft.

e152

JOURNAL OF SURGICAL RESEARCH: VOL. 168, NO. 1, JUNE 1, 2011

with infected prosthetic grafts or primary arterial infections. Similarly, the presence or absence of sepsis seems to be an important factor in patients with cardiovascular infections [1,2]. Investigators have recently emphasized the importance of SIRS and its relationship to outcomes in patients with sepsi [1–3]. Patients with SIRS have two of the following criteria: 1. 2. 3. 4.

Temperature < 37 C or > 38 C. Heart rate > 90. Respiratory rate > 20 or a pCO2 < 32 mmHg. A white blood cell count < 4000 or > 12,000.

Among their patients with infected aneurysms, Ihaya et al. reported a mortality rate of 75% among those with SIRS compared with a 20% mortality rate in those without SIRS [1]. Eleven of our patients met these criteria for SIRS. One of these patients experienced early graft failure, four patients died within 30 d, and another three died within 90 d (63.6% mortality rate overall). The mortality rates that we report are indeed quite high. However, a significant number of our patients presenting with SIRS had aorto-bronchial, aorta-intestinal fistulae, or had rupture of their infected aneurysms. In contrast to this, we had no deaths and no allograft complications among the eight patients with infections not involving the thoracic or abdominal aorta. Only one of our patients developed subsequent late failure of an allograft. Even though we have experienced only one late allograft failure, other series have reported a higher incidence of long-term complications with CPAs. In reviewing the literature on long-term complications, we focused only on reports in which CPAs were used as opposed to fresh allografts. Nevelsteen et al. reported a 31% incidence of subsequent allograft abnormalities including mild dilatation, mural thrombus formation, and concentric calcification of the allograft [4]. Among these 90 cases with CPAs used in the treatment of infected prosthetic grafts, there were seven cases of graft dilatation on late follow-up and four cases with graft stenosis. Six patients also developed graft thrombosis. Leseche et al. reported a 17% incidence of allograftrelated problems including aneurysmal dilatation of the graft or graft thrombosis [9]. In their review of a multi-centered registry of CPAs, Noel et al. reported significant complications as well [16]. We found only one study that compared the use of CPAs and prosthetic graft infections to another group of patients with prosthetic graft infections treated by alternative means such as extra-anatomic bypasses or in situ reconstruction [8]. These authors reported a 17.9% incidence of graft-related morbidity among those patients with CPAs compared with 57.1% incidence graft-related morbidity in the other group. These

authors also found the CPAs to be resistant to reinfection, thrombosis, or aneurysmal dilatation. The primary limitation of this study is that not all of our patients were compliant in follow-up despite our best efforts in encouraging them to do so. Because not all patients returned for follow-up, it is certainly possible that there were more graft complications than we have reported. One would assume, however, that these patients would have been referred to us for any complications related to their allografts. Also, we did not have a comparable group of patients treated by other means. CONCLUSIONS

We believe that CPAs are a reasonable option in treating patients with either graft infections or native arterial infections. Certain subgroups, such as those with aorto-intestinal fistulae or aortobronchial fistulae have a high mortality rate. CPAs may have distinct advantages in treating these patients compared with other treatment modalities such as in situ reconstructions in shorter operative times. The patients presenting with overwhelming sepsis have a very high mortality rate. Long-term complications of CPAs are uncommon. Surgeons implanting these grafts should feel confident about the long-term durability of these allografts. ACKNOWLEDGMENTS Dr. Fehrenbacher and Dr. McCready are members of the Physician Advisory Board for CryoLife.

REFERENCES 1. Ihaya A, Chiba Y, Kimura T, et al. Surgical outcome of infected aneurysms of the abdominal aorta with or without SIRS. Cardiovasc Surg 2001;9:436. 2. Fillmore AJ, Valentine RJ. Surgical mortality in patients with infected aneurysm. J Am Coll Surg 2003;196:435. 3. Muckhart DJ, Bhagwanjoes S. Amer Coll of Chest Physicians/ Society of Critical Care Medicine Consensus Conference Committee. Definitions of the systemic inflammatory response syndrome and allied disorders in relation to critically injured patients. Crit Care Med 1997;25:1789. 4. Nevelsteen A, Feryn T, LaCroix V, et al. Experiences with CPAs in the treatment of prosthetic graft infections. Cardiovasc Surg 1998;6:378. 5. Verhelst R, Lacroix V, Vraux H, et al. Use of cryopreserved homografts for management of infected prosthetic grafts: A multicentric study. Ann Vasc Surg 2000;14:602. 6. McCready R, Bryant MA, Divelbiss J, et al. Arterial infections in the new millennium: An old problem revisited. Ann Vasc Surg 2006;20:590. 7. McCready R, Bryant MA, Divelbiss J, et al. Case study: Chronic femero-popliteal prosthetic graft infections with exposed graft. Vasc Endovasc Surg 2009;43:291. 8. Brown KE, Heyer K, Rodriguez H, et al. Arterial reconstruction with cryopreserved human allografts in the setting of infection: A single-center experience with midterm follow-up. J Vasc Surg 2009;49:660.

MCCREADY ET AL.: LONG-TERM RESULTS WITH CRYOPRESERVED ARTERIAL ALLOGRAFTS (CPAS) 9. Leseche G, Castier Y, Petit MD, et al. Long-term results of cryopreserved allograft reconstruction in infected prosthetic grafts and mycotic aneurysms of the abdominal aorta. J Vasc Surg 2001; 34:616. 10. Vogt PR, Pfammatter T, Schlumpf R, et al. In situ repair of aortobronchial, aorto-esophageal, and aorto-enteric fistulas with cryopreserved homografts. J Vasc Surg 1997;26:11. 11. Castier Y, Francis F, Cerceau P, et al. Cryopreserved arterial allograft reconstruction for peripheral graft infections. J Vas Surg 2005;41:30. 12. Vogt PR, Vonsegesser LK, Goffin Y, et al. Eradication of aortic infections with the use of cryopreserved arterial homografts. Ann Thorac Surg 1996;62:640.

e153

13. Kieffer E, Gomes D, Chiche L, et al. Allograft replacement for infrarenal aortic graft infections: Early and late results in 179 patients. J Vasc Surg 2004;39:1009. 14. Zhou W, Lin PH, Bush RL, et al. In situ reconstruction with cryopreserved arterial allografts. Tex Heart Instit J 2006;33:14. 15. Vogt PR, Brunner-LaRocca HP, Lachat M, et al. Technical details with the use of cryopreserved arterial allografts for aortic infections: Influence on early and midterm mortality. J Vasc Surg 2002;35:80. 16. Noel AA, Gloviczki P, Cherry KJ Jr, et al. Abdominal aortic reconstruction in infected fields: Early results of the U.S. Cryopreserved Aortic Allograft Registry. J Vasc Surg 2002; 35:847.