Abstract

Methods: Between December 1985 and January 2011, 42 consecutive patients (32 males, 10 females; mean age 39.0±13.3 years; range 16 to 67 years) were retrospectively analyzed in terms of surgical findings and operative procedures. The blood cultures were positive in 25 patients (59.5%) and the most commonly identified microorganism was streptococcus (n=15, 35.7%). Fourteen patients (33.3%) had a medical history of previous cardiac surgery and 13 (31.0%) had prosthetic valve endocarditis. The mean duration of follow-up were 7.9±4.4 years (range 0.1 to 18.2 years).

Results: All patients underwent a total of 64 surgical procedures. The most commonly performed procedure was aortic valve replacement with 26 patients (61.9%), followed by aortic root replacement in 15 (35.7%) and primary repair of periprosthetic leakage in one patient (2.4%). Nine patients (21.4%) had concomitant procedures for the mitral valve. In-hospital mortality was seen in nine patients (21.4%). Postoperatively seven patients had (16.7%) low cardiac output, six had (14.3%) heart block, however, only two of them required permanent pacemaker. The actuarial survival rates at one, five and 10 years were 80.0±6.3%, 69.9±7.3% and 64.9±8.3%, respectively.

Conclusion: Although surgery for aortic valve endocarditis with annular involvement has high in-hospital mortality rate, long-term survival is satisfactory for surviving patients.

Methods

Twenty-nine patients (69.0%) presented with NVE and 13 (31.0%) with PVE, with early PVE cases being excluded from the analysis. No significant differences were seen in the average ages of the patients with NVE and PVE (38.8±12.7 vs. 39.5±14.9, respectively; p=0.415). Vegetations on the mitral prosthetic valve were detected in three cases while mitral valve periprosthetic leakage was present in two others. In addition, 20 patients (33.3%) had a previous history of cardiac surgery, and four had undergone two cardiac operations. Furthermore, 11 patients had been operated on for aortic valve procedures and four for double valve procedures. The remaining patient had aortic reconstruction but no replacement. The preoperative characteristics are summarized in Table 1. We did not perform coronary angiography on any of the patients in order to avoid embolic complications.

Table 1: Preoperative characteristics

The diagnosis of infective endocarditis was made according to the Duke criteria.[5] All of the patients were examined by either transthoracic (TTE) or transesophageal (TEE) echocardiography, which revealed valvular vegetation, annular abscesses, aorticoatrial communication, or periprosthetic leakage. Intraoperatively, annular involvement was considered when an abscess penetrated into the valvular annulus or the adjacent myocardial structures or when the vegetations attacked the aortic annulus or adjacent structures. Gross vegetations were detected in 12 patients (28.6%) preoperatively.

Indications for surgery and valve choice
The patients were scheduled for surgery when the diagnoses were made, and 22 (52.4%) were operated on in the active phase of the infection. The indications for emergency surgery were big (>10 mm), mobile vegetations on the aortic valve, acute leaflet rupture and cardiac decompensation, a periannular extensive abscess with an intracardiac fistula, and prosthetic valve dysfunction. In 20 patients (47.6%), the operation was performed after the antibiotic treatments were completed and the patient was stabilized. The valve choice was primarily made according to the degree of destruction in the annular area. Prosthetic materials were used when the aortic root was not extensively destroyed by the infectious process, with homografts and xenografts being used when available.

Microbiological studies
In addition, the blood cultures were positive in 25 patients (59.5%), and the most commonly identified microorganisms were various strains of streptococcus (35.7%). The results of the microbiological studies can be seen in Table 2.

Table 2: Microbiological study results

Operative technique
All patients underwent moderate (28 °C) hypothermic cardiopulmonary bypass (CPB) via bicaval cannulation of either the ascending aorta (n=39) or the femoral artery (n=3). Isothermic blood cardioplegic solution was also administered by the retrograde route during aortic cross-clamping.

The most commonly performed procedure was aortic valve replacement in 26 patients (61.9%) followed by aortic root replacement in 15 (35.7%) and primary repair of periprosthetic leakage in one (2.4%). For eradication of aortic valve endocarditis, aortic annular skeletonization was performed. All infected and necrotic tissue around the annulus, and when present, within the abscess and fistula between the ventriculoarterial junction and the sinotubular junction, were resected. In addition, all vegetations were also removed. When necessary, a resected annular area was covered with a glutaraldehyde-treated autologous pericardial patch sutured to firm, fibrous tissue for a secure anastomosis or valve implantation. The approaches to the patients with fistulas have been previously reported.[6,7] The list of procedures can be seen in Table 3.

Table 3: Procedures

The mitral valve was explored in all cases, with appropriate procedures being performed in nine of the patients. In addition, 15 had aortic root replacements, one underwent replacement of the prosthesis with aortic root enlargement, and another underwent concomitant subaortic discrete membrane resection (Table 3). The average aortic cross-clamp and total perfusion times were 111.0±30.8 (range 46-188) and 146.0±63.7 (range 65-411) minutes, respectively.

Follow-up
All patients received at least four weeks of postoperative antibiotic therapy. Broad -range antibiotics (vancomycine and aminoglycosides) were preferred for the culture- negative cases while the other patients were treated according to their antibiograms. The patients were followed up in the outpatient clinic of our hospital for a mean duration of 7.9±4.4 years (0.1-18.2), for a total of 244.2 patient-years.

Statistical analysis
All statistical analyses were performed using the SPSS version 16.0 statistical software package (SPSS Inc., Chicago, IL, USA). The continuous variables were expressed as mean ± standard deviation (SD), and the ranges were also calculated. Furthermore, the discrete variables were expressed as frequencies and percentages. Comparisons of the discrete variables were made via a chi-square test, and the survival, freedom from recurrence, and freedom from reoperation analyses were conducted using the Kaplan- Meier estimate. The survival comparisons were made with a log-rank test, and p values of less than 0.05 were accepted as statistically significant differences.

Results

Morbidity
Postoperative fever was seen in 14 patients (33.3%), and three of these had no fever prior to the surgery. Complete heart block was present in six patients (14.3%) postoperatively, but only two of them (4.8%) required the implantation of a permanent pacemaker. In addition, two of these patients had PVE and four had aortic root procedures. Moreover, only one of the patients with postoperative heart block who had undergone a redo aortic valve replacement (AVR) and aortic root enlargement died in the early postoperative period. Furthermore, no mortality occurred in the patients who required a permanent pacemaker. Renal dysfunction was present in 12 patients (28.6%), with four (9.5%) requiring dialysis. Pulmonary morbidity was seen in six patients (14.3%), and two others (4.8%) had postoperative cerebrovascular events. However, these two patients did not have any septic emboli preoperatively. One of them had a recurrence and a reoperation during the hospitalization period, and he died on the 45th p ostoperative d ay. T he p atients w ith PVE and NVE were compared regarding postoperative morbidity, and although the absolute frequencies varied, they were not statistically significant (61.5% vs. 31.0%, respectively; p=0.063).

Follow-up
Of the 33 patients who survived, 31 completed the follow-up process. Two patients were lost to follow-up, and one of these had been operated on 26 years earlier. Additionally, four patients (12.9%) had mortality after discharge, and one of them died suddenly one month after leaving the hospital because of congestive heart failure. He had been categorized as New York Heart Association (NYHA) Class 4 preoperatively, but his left ventricular function was normal. His preoperative cultures were also negative. One other patient had prosthetic valve endocarditis along with streptococcal growth in his cultures. He had previously undergone two operations before having infective endocarditis, with prosthetic aortic valve implantation being initially performed followed later by aortic root replacement. His cardiac functions had also been normal. Unfortunately, he had a recurrence of infective endocarditis and failed to respond to therapy. He died six months after being discharged. In addition, one of the long-term survivors died from a stroke in his ninth year of follow-up, and another had prosthetic valve endocarditis along with postoperative renal dysfunction with a need for dialysis. He died in the fourth year of follow-up because of chronic renal failure. The actuarial survival for one, five, and 10 years was 80.0±6.3%, 69.9±7.3%, and 64.9±8.3%, respectively. When the patients with PVE and NVE were compared, the difference was statistically significant (p=0.046). The actuarial survival for patients with PVE at one, five, and 10 years was 61.5±13.5%, 46.2±13.8%, and 46.2±13.8%, respectively, whereas it was 81.5±7.5%, 81.5±7.5% and 73.3±10.2% for patients with NVE at the same intervals.

A recurrence of infection occurred in three cases, and one of these was summarized earlier in the article. The other two cases had relapses of infection before being discharged and were taken for reoperations. One was a second redo case who had undergone aortic valve replacement, and he had his prosthesis replaced with aortic root enlargement. Primary repair of the periprosthetic leak attributable to the infective endocarditis was also performed on this patient. After recurrence, this patient underwent a reoperation in which the prosthesis was replaced, but he died of sepsis on the 12^th postoperative day. The other patient had an native valve disease. He underwent a homograft replacement, but there was a functional failure that resulted in low cardiac output. A reoperation was performed in the first postoperative month, but during the postoperative course, sepsis developed, causing mortality on the 45th postoperative day. No other patients required reoperations apart from these three cases. Although none of the patients who received biological grafts had recurrences, the difference was not statistically significant (0% vs. 10.7%, respectively; p=1.000). Furthermore, the freedom from recurrence from infection at the first year was 91.8±4.6%, and this remained stable throughout the follow-up period. Moreover, the freedom from reoperation was 95.0±3.4% in the first year, and this also remained stable.

Discussion

The use of biological materials for patients with aortic PVE is recommended due to the low reinfection rates caused by the increased penetration of antibiotics in these tissues and the presence of viable cells.[14] Hagl et al.[15] reported their experience regarding the use of a mechanical prosthesis in patients with PVE of the aortic valve and identified a mortality rate of 11% in 28 patients. They only had one recurrent case (4%) that died during the reoperation, and this happened at the postoperative third month. In addition, they reported a five-year survival rate of approximately 80%. In another study by David et al.,[16] they reported a 60% survival rate after five years. Our results are compatible with these reports. Although not statistically significant, none of the patients with biological grafts experienced a recurrence. This may be because of the low number of biological grafts available, and the increased use of biological grafts may decrease the recurrence rate. Knosella et al.[12] also reported significant differences between the allograft and prosthetic groups in their study for both in-hospital and long-term mortality.

Although the total survival rate in our study was compatible with the findings in the study by Hagl et al.,[15] the PVE results and the number of cases with NVE were lower in our series than theirs. Two possible explanations may account for this contradiction. One reason may be the lower number of cases in our series. We think this is important because the difference in morbidity rates was not significant even though the absolute frequency was about two times higher in the NVE cases. The statistical power could also have been lower. In addition, the high rate of morbidity in our series could also be responsible. As previously pointed out, the morbidity rates were higher with PVE, which may be significant. To verify this, a regression analysis could be performed on a larger number of patients. David et al.[16] reported that a preoperative state of shock and the involvement of both the aortic and mitral annuli were associated with mortality, and our findings related to long-term survival were compatible with that study,[16] with the differences between NVE and PVE being strikingly similar. Knosalla et al.[12] also reported that preoperative shock and sepsis were predictors of mortality.

The David et al.[16] study also reported that about a third of their patients with active infective endocarditis had a paravalvular abscess, and our rates (24%) were compatible with theirs. They also determined that only 60% of their cases were diagnosed preoperatively. In our series, most of the cases were diagnosed intraoperatively as an annular involvement.

Secondary involvement of the mitral valve is important in patients with an aortic root abscess. Siniawski et al.[17] found that approximately 25% of their cases had secondary mitral valve disease. In our report, we determined that 23.8% of the patients underwent a mitral valve procedure because of the infectious process. The same authors also reported an in-hospital mortality rate of 26.4% in the patients for whom double valve surgery was performed. Our series found nine patients who had concomitant mitral valve surgery (Table 3), and three of those (33.3%) suffered in-hospital mortality.

One significant difference between our report and others that have been published[12,15,16] was the presence of the growth of microorganisms in the blood cultures. As shown in Table 2, around 40% of the cases were culture-negative, and the most frequently isolated bacteria was streptococcus, which contrasted with the staphylococcal predominance in the other reports. The differences between these studies and ours may be due to two reasons. One is the high rate of culture-negative cases in our study, possibly caused by the inappropriate handling of the blood cultures and the incorrect timing of the blood sampling. In most cases, blood samples are drawn when the patients are on antibiotics. The second reason may be the improper treatment of patients who are infected with streptococci, which could have led to more serious consequences.[16]

Our study had several drawbacks. First, we had a very high rate of culture-negative cases, and this was discussed in detail in a previous analysis.[18] Thus, our findings regarding the microorganisms must be viewed with caution. In addition, we had a low number of patients which did not allow us to perform a multivariable analysis. Some also might argue that the low rates of biological materials used in our study could also be a negative. Finally, one of the major limitations of this study was that the operations took place over a 15-year period, and the operative and medical therapies evolved markedly over this time. Although conducting a retrospective study over this length of time was not ideal, it was necessary in order to make inferences regarding survival and followup. In spite of these drawbacks, we believe that our report is relevant because it consisted of patients who were operated on at a single center over an extended period of time and because there was a significant postoperative follow-up period.

Conclusion

Declaration of conflicting interests
The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

Funding
The authors received no financial support for the research and/or authorship of this article.

References

1) Hill EE, Herijgers P, Herregods MC, Peetermans WE. Evolving trends in infective endocarditis. Clin Microbiol Infect 2006;12:5-12.

2) d’Udekem Y, David TE, Feindel CM, Armstrong S, Sun Z. Long-term results of operation for paravalvular abscess. Ann Thorac Surg 1996;62:48-53.

3) Ergin MA, Raissi S, Follis F, Lansman SL, Griepp RB. Annular destruction in acute bacterial endocarditis. Surgical techniques to meet the challenge. J Thorac Cardiovasc Surg 1989;97:755-63.

4) Aranki SF, Santini F, Adams DH, Rizzo RJ, Couper GS, Kinchla NM, et al. Aortic valve endocarditis. Determinants of early survival and late morbidity. Circulation 1994;90:II175-82.

5) Durack DT, Lukes AS, Bright DK. New criteria for diagnosis of infective endocarditis: utilization of specific echocardiographic findings. Duke Endocarditis Service. Am J Med 1994;96:200-9.

6) Bozbuga N, Erentug V, Erdogan HB, Kirali K, Ardal H, Tas S, et al. Surgical treatment of aortic abscess and fistula. Tex Heart Inst J 2004;31:382-6.

7) Kirali K, Omeroglu SN, Mansuroglu D. Aortic root abscess with fistula formation into right ventricular myocardium. Türk Kardiyol Dern Arş 2000;28:647-9.

8) Wallace AG, Young WG Jr, Osterhout S. Treatment of acute bacterial endocarditis by valve excision and replacement. Circulation 1965;31:450-3.

9) Yankah AC, Pasic M, Klose H, Siniawski H, Weng Y, Hetzer R. Homograft reconstruction of the aortic root for endocarditis with periannular abscess: a 17-year study. Eur J Cardiothorac Surg 2005;28:69-75.

10) Shimizu H, Takahashi T, Yamazaki M, Hasegawa N, Shin H, Yozu R. Aortic root re-replacement with cryopreserved aortic homograft in a patient with active composite valvegraft endocarditis. Keio J Med 2008;57:111-4.

11) Jashari R, Goffin Y, Van Hoeck B, Vanderkelen A, du Verger A, Fan Y, et al. Belgian and European experience with the European Homograft Bank (EHB) cryopreserved allograft valves.--assessment of a 20 year activity. Acta Chir Belg 2010;110:280-90.

12) Knosalla C, Weng Y, Yankah AC, Siniawski H, Hofmeister J, Hammerschmidt R, et al. Surgical treatment of active infective aortic valve endocarditis with associated periannular abscess--11 year results. Eur Heart J 2000;21:490-7.

13) David TE. Surgical management of aortic root abscess. J Card Surg 1997;12:262-6.

14) Fischlein T, Schütz A, Uhlig A, Frey R, Krupa W, Babic R, et al. Integrity and viability of homograft valves. Eur J Cardiothorac Surg 1994;8:425-30.

15) Hagl C, Galla JD, Lansman SL, Fink D, Bodian CA, Spielvogel D, et al. Replacing the ascending aorta and aortic valve for acute prosthetic valve endocarditis: is using prosthetic material contraindicated? Ann Thorac Surg 2002;74:S1781-5.

16) David TE, Regesta T, Gavra G, Armstrong S, Maganti MD. Surgical treatment of paravalvular abscess: long-term results. Eur J Cardiothorac Surg 2007;31:43-8.

17) Siniawski H, Grauhan O, Hofmann M, Pasic M, Weng Y, Yankah C, et al. Aortic root abscess and secondary infective mitral valve disease: results of surgical endocarditis treatment. Eur J Cardiothorac Surg 2005;27:434-40.

18) Polat A, Tuncer A, Tuncer EY, Mataraci I, Aksoy E, Donmez AA, et al. Surgical treatment of culture-negative aortic infective endocarditis. Ann Thorac Surg 2012;93:44-9.