| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cardiovascular and Thoracic Surgery, Cardiothoracic Section, Aarhus University Hospital Skejby, Aarhus, Denmark
b Department of Clinical Microbiology, Aarhus University Hospital Aalborg, Aalborg, Denmark
c Department of Cardiothoracic Anesthesia and Intensive Care, Center for Cardiovascular Research, Aarhus University Hospital Aalborg, Aalborg, Denmark
Accepted for publication August 7, 2009.
* Address correspondence to Dr Modrau, Department of Cardiovascular and Thoracic Surgery, Cardiothoracic Section, Aarhus University Hospital Skejby, Brendstrupgaardsvej 100, 8200 Aarhus N, Denmark (Email: modrau{at}mail1.stofanet.dk).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods: A retrospective observational study was performed at Aalborg Hospital, Aarhus University Hospital, Denmark, from January 1999 through November 2006. Included were all 83 of 4222 cardiac surgical patients with microbiologically documented DSWI requiring surgical revision. Various potential risk factors in patients with Candida DSWI were compared with those of patients with non-Candida DSWI. We compared markers of morbidity, in-hospital mortality, and 1-year mortality to evaluate the prognosis of the disease.
Results: DSWI developed in 2% of all patients, of whom, 17 (20.5%) had Candida DSWI, and 66 (79.5%) had non-Candida etiology. Candida was the primary causative organism in 11 of 17 Candida DSWI cases. No Candida DSWI was found during the first 3 years of the study. In-hospital and 1-year mortality were doubled in patients with Candida DSWI compared with patients with non-Candida DSWI. Candida DSWI was associated with significantly longer stay in the intensive care unit and need of prolonged mechanical ventilation. Risk factors for Candida etiology were Candida colonization in tracheal secretions or urine and reoperation before diagnosis of DSWI.
Conclusions: Candida was a frequent causative agent of DSWI in our series and was associated with a very high morbidity and mortality. Cardiothoracic patients on mechanical ventilation when colonized with Candida were identified as a high-risk population for subsequent development of Candida DSWI.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
A substantial number of DSWIs caused by Candida spp at our institution prompted us to conduct this systematic retrospective analysis. Environmental hygiene investigations of personnel, operation rooms, the intensive care unit (ICU), and ward facilities during the study period failed to prove any signs of Candida contamination. The aim of the present study was to determine the overall incidence and risk factors for development of Candida DSWI. Our objective was to compare the prognosis of Candida with non-Candida DSWI.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Study Design and Data Collection
A total of 4222 adult patients undergoing sternotomy for cardiac operations at Aalborg Hospital, Aarhus University Hospital, Denmark, were eligible during the study period from January 1999 through November 2006. The procedures included acute and elective surgery, on-pump and off-pump coronary artery bypass grafting, valve replacement, valve repair, and interventions on the ascending aortic artery.
During the study period, all patients received cefuroxime (1.5 g intravenously [IV], thrice daily) and gentamicin (240 mg IV, once daily) as antibiotic prophylaxis. Antibiotics were administered at the induction of anesthesia and were continued for 48 hours. If reoperation for other reasons than DSWI was required, the prophylactic antibiotic treatment was continued for 5 days. We did not to include patients from December 2006 because the prophylactic antibiotic regimen was changed at that time.
Patients with DSWI were identified through the specific procedure code for reoperation for deep infection after cardiac operations (FWC00) and through the clinical microbiology database of our institution. The identified cases of DSWIs were verified by record review of all patients. Patients were divided into subgroups with and without Candida etiology.
The standard antifungal treatment was fluconazole (400 mg IV, once daily) in all cases of C albicans DSWIs. Amphotericin B (0.6 to 1.0 mg/kg IV, once daily) was given in patients with non-C albicans infections or who did not respond to the first-line treatment. Antifungal treatment continued for at least 1 month.
The standard approach of surgical treatment in case of DSWI was early closure with antibiotic wound irrigation during the first 3 years of the study period. During the last 5 years, these patients were treated with vacuum-assisted closure technique. Final closure was completed with or without plastic reconstruction at the surgeon's discretion. Patients with discrete signs of infection were closed primarily after ample débridement and irrigation.
The following data were abstracted from the medical records using European System for Cardiac Operative Risk Evaluation (EuroSCORE) definitions: diabetes, chronic obstructive pulmonary disease, extracardiac arteriopathy, ejection fraction, past cardiac operation, urgent or emergency operation, and preoperative dialysis [8]. We also registered obesity (body mass index > 30 kg/m2), smoking, preoperative antibiotic therapy, alcohol abuse, reoperation for other causes before DSWI, length of mechanical ventilation, length of stay in the ICU, and duration of postoperative hemodialysis.
Mediastinal tissue cultures were taken in all patients at the time of reexploration for DSWI. Cultures of tracheal secretions and urine were performed in all patients with clinically suspected infection before diagnosis of DSWI. All bacteriologic culture results were obtained from the clinical microbiology database of our institution. The remaining data were retrieved from the population-based Danish National Patient Register that contains the unique 10-digit civil registration number of every Danish citizen, including age and gender, vital status, admission and discharge dates, date and kind of operation, and discharge diagnoses.
Definitions of DSWI and Outcomes
DSWI was defined as deep infection involving retrosternal tissue or the sternal bone, or both, directly observed by the surgeon and confirmed by positive culture of a specimen from a bone or deep tissue biopsy or pericardial fluid obtained during reexploration. Superinfections were defined as appearance of new pathogens in the above specimen more than a week after the primary diagnosis of DSWI.
The primary outcome measured was all-cause mortality at the time of discharge and after 1 year. Secondary outcome variables assessed for postoperative course were length of stay in the ICU, mechanical ventilation, total hospital stay, and postoperative hemodialysis.
Statistical Analysis
Descriptive statistics included median (range) and number (percentage) of patients when appropriate. Differences between the two groups of patients were tested for statistical significance using Wilcoxon rank sum test (nonnormally distributed data) or Fisher exact test (categoric variables). A value of p < 0.05 was considered to be statistically significant. Comparisons were made between patients with complete data for the given analysis. We computed risk ratios (RR) with 95% confidence intervals (CI) to compare the two groups for the different risk factors using Poisson regression with robust variances [9]. All analyses were calculated with Stata 9.2 software (StataCorp, College Station, TX).
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Of the 83 studied patients with DSWI after median sternotomy for a cardiac operation, 66 (79.5%) had non-Candida etiology and 17 (20.5%) had Candida etiology. The overall incidence of Candida DSWI in our cohort was 0.4%. Candida represented the primary causative organism in 11 patients, and 6 had Candida superinfections.
No Candida DSWI was found during the first 3 years of the study. During the last 5 years of the study, Candida DSWI occurred with a frequency of 14% to 42% of all DSWIs, with no identifiable pattern. Fourteen Candida DSWI patients were treated with vacuum-assisted closure technique until secondary closure, 2 of whom had recurrences with reopening of the sternum. One patient was closed primarily. Two patients died, at 1 and 3 days after diagnosis.
As reported in Table 1, there was no significant difference in distribution of the potential risk factors for development of Candida DSWI between the two groups. However, there was a substantial difference in mortality between the two groups. In-hospital all-cause mortality was 35% in patients with Candida DSWI vs 15% in patients with non-Candida etiology (RR, 2.28; 95% CI, 0.96 to 5.44; p = 0.061). The 1-year all-cause mortality was 41% in patients with Candida DSWI vs 23% in patients with non-Candida etiology (RR, 1.94; 95% CI, 0.83 to 4.55, p = 0.124). The total in-hospital and 1-year all-cause mortality for all DSWI patients was 19% and 27%, respectively. The characteristics of the postoperative course in both groups are detailed in Table 2. Candida DSWI was associated with significantly longer ICU stay and need of prolonged mechanical ventilation.
|
|
The most commonly isolated pathogen in primary DSWI was Staphylococcus epidermidis (50% of isolates), followed by S aureus (16%) and gram-negative rod-shaped microbes (12%). Other causative pathogens were Candida spp (10%), Streptococci (8%), and anaerobic bacteria (4%), and 19 (23%) primary infections were polymicrobial.
Superinfections occurred in 35 patients (42%), of which 14 (17%) were polymicrobial. Among superinfections, gram-negative rod-shaped microbes were the most prevalent pathogens (37% of isolates). The most frequent gram-negative rod-shaped microbes were Pseudomonas aeruginosa and Escherichia coli, each accounting for 8% of isolates, followed by Klebsiella and Proteus spp, each according for 7% of isolates. Other common causative pathogens in superinfections were S epidermidis (30% of isolates), and Candida spp (10%). In our study, all Candida spp isolated were C albicans, with the exception of one fulminant primary infection with C lusitaniae.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
In our study, we found remarkable differences in mortality between the patients with and without Candida etiology for DSWI. In-hospital mortality was more than doubled in patients with Candida DSWI compared with non-Candida DSWI, and the substantial difference in mortality persisted after 1 year. The major limitation of this study and other retrospective observational studies is the difficulty to account adequately for the confounding effect of the underlying illness.
Our total in-hospital and 1-year mortality rates for DSWI are within the ranges of early and late mortality rates reported in other recent series [3, 5, 14]. However, another recent Scandinavian study reported a substantially lower 30-day mortality of 1%, and 1-year mortality of 7.2% [6]. As in our study, the authors were able to ascertain reliable mortality data from a population-based national registry. The discrepancy can partly be explained by the inclusion of 22% of patients with negative cultures, a relative high incidence of DSWI of 3.2%, and no patients with Candida etiology.
Patients with Candida DSWI stayed significantly longer in the ICU and required mechanical ventilation 10 times longer than patients with DSWI of other etiology. The length of mechanical ventilation has been proposed as an independent predictor of candidemia and candidemia-related death in cardiothoracic ICU patients [13]. Nevertheless, uncertainty remains whether these findings more likely reflect the patients' severity of disease.
The rate of DSWI in our study is within the range of 1.4% to 3.6% reported in other recent observational studies [2, 6, 7, 14, 15]. Candida as the cause of DSWI is considered to represent a rare clinical entity. There are few reports of nosocomial outbreaks [16–18]. DSWI after sternotomy due to Candida spp has an incidence of 0% to 7.5% in the most recently reported series [1, 2, 4–7]. In contrast, we observed Candida etiology in every fifth patient with DSWI. Several explanations could be proposed to explain the high incidence of Candida DSWI in our study.
First, a few previous series involved a large proportion of patients treated with closed antibiotic wound irrigation [19–21]. These series report incidences of Candida DSWI of up to 15%. However, Candida DSWI did not occur in our series during the first 3 years, where closed antibiotic wound irrigation was applied.
Second, our high incidence of Candida DSWI could reflect a high proportion of patients susceptible to opportunistic infections in the data. However, our analysis of potential risk factors for development of Candida DSWI did not show any significant difference in distribution between the two groups (Table 1).
Third, early hospital discharge and the lack of follow-up might lead to the underestimation of the incidence of DSWI with late and insidious onset. In our study, data retrieval from the population-based national registry permitted registration of all cases of rehospitalization within the border of the country.
Finally, the low incidence of reported Candida DSWI may be due to the propensity to overlook the pathogenic potential of Candida spp as contamination or superinfection. In our study, Candida represented the primary causative organism in most patients. In addition, the time until diagnosis of DSWI was similar in Candida compared with non-Candida etiology. Our findings suggest Candida, isolated from deep mediastinal tissue at operation, to represent a true pathogen.
The high incidence, and high mortality associated with systemic Candida infections has focused recent interest on preventative strategies. A recent Cochrane Review concluded that antifungal prophylaxis should be considered in nonneutropenic critically ill patients [22]. Because the generation of resistance to antifungal agents remains a major concern of prophylactic treatment, the accurate identification of patients at increased risk is crucial. Many risk factors for fungal infections, such as colonization, length of stay, use of parenteral nutrition, antibiotics, central catheters, and abdominal operations have been defined in different patient populations in the ICU setting [23]. In addition, Pittet and coworkers [24] demonstrated Candida colonization always preceded systemic infection with genotypically identical Candida spp strain in a surgical and neonatal ICU setting.
The current study confirmed that Candida colonization in tracheal secretions or urine was associated with a significantly increased risk of Candida DSWI. As a limitation, cultures of tracheal secretions and urine were performed more often in patients with Candida etiology than in patients with non-Candida etiology. However, the influence of this potential selection bias is considered negligible. The probability of a positive culture is presumed low in patients in whom no signs of infection had prompted microbiologic investigations.
From our data, we conclude that cardiothoracic patients on mechanical ventilation when colonized with Candida in tracheal secretions or urine are at high-risk for subsequent development of Candida DSWI. Future studies are required to validate whether these patients should be considered for preemptive antifungal treatment. We also found reoperation for other causes before the diagnosis of infection was a risk factor for Candida DSWI. This finding could reflect the administration of prophylactic antibiotic treatment for the extended period of 5 days in these patients.
Our study is limited by its retrospective design, with a relatively small number of patients in the subgroups compared. The initial cohort was retrieved from a single cardiothoracic center. We validated all data by crosschecking two clinical databases and medical record review to diminish the risk of underreporting. The results of our study were in accordance to previous findings. Thus, we consider our findings as robust and applicable in the clinical setting of critically ill cardiothoracic patients.
In conclusion, this investigation has important clinical implications. We demonstrate that Candida DSWI is an important emerging clinical entity associated with significantly increased morbidity and mortality compared with DSWIs of other etiology. Cardiothoracic patients on mechanical ventilation when colonized with Candida were identified as a high-risk population for subsequent development of Candida DSWI.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  A. T. Culliford Invited Commentary Ann. Thorac. Surg., December 1, 2009; 88(6): 1909 - 1909. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
