- Email: [email protected]
Postoperative Mediastinitis in Children: Improvement of Simple Primary Closed Drainage
Christine Anslot, MD, Sylvie Hulin, MD, and Yves Durandy, MD Intensive Care Unit of Pediatric Cardiac Surgery, Institut Jacques Cartier, Massy, France
Background. Mediastinitis is a significant cause of postoperative morbidity. In 1989, we proposed simple primary closed drainage as a new treatment. Our goal is to describe improvements made to the original technique. Methods. After wound debridement, infected areas were drained with Redon catheters connected to strong negative-pressure drainage bottles. Mediastinal effluents were cultured every day, and the catheters were withdrawn when the effluent culture was negative for microorganisms. Patients were classified into three groups: isolated mediastinitis (group 1), mediastinitis associated with endocarditis (group 2), and mediastinitis associated with other organ failure (group 3). Results. Sixty-four patients were treated during a 10year period: 15 neonates, 33 infants, and 16 children. Group 1 consisted of 40 patients. The time to mediastinal sterilization was 4 days (range, 1 to 14 days), and the antibiotic course was 11 days (range, 7 to 28 days), with a hospital stay of 13 days (range, 10 to 30 days). No deaths
occurred in this group. Group 2 consisted of 7 patients. The time to mediastinal sterilization was 8 days (range, 3 to 10 days), and the antibiotic course was 30 days (range, 26 to 37 days), with a hospital stay of 37 days (range, 20 to 54 days). One patient in group 2 did not survive. Group 3 consisted of 17 patients. The time to mediastinal sterilization was 6 days (range, 1 to 10 days), and the antibiotic course was 15 days (range, 10 to 31 days), with a hospital stay of 20 days (range, 18 to 36 days). Two patients in group 3 did not survive. None of the deaths was directly related to mediastinitis, as the mediastinum was sterile in all 3 patients before death. Conclusions. This simple treatment was efficient and reliable in achieving mediastinal sterilization. In addition, short antibiotic courses decreased restraint, which is poorly tolerated in pediatric patients.
M
Material and Methods
ediastinitis is a major complication ensuing pediatric cardiac surgery. Its treatment remains controversial. The first part of therapeutic management is always surgical revision, which is completed with one of the following procedures: closed irrigation, an old technique first described in 1963 [1] and still used [2]; delayed or primary muscle flaps [3–5]; vacuum-assisted closure, first described in adults but also used in infants [6, 7]; or primary closure with short-term drainage [8]. In 1989 we proposed simple primary closure with drainage through a Redon catheter [9], which was subsequently practiced by others [10 –12]. Since then we have learned from much experience that it is possible to shorten the drainage time and course of chemotherapy. This strategy also demonstrates benefits in terms of decreasing the physical and psychological effects of hospital stays and intravenous therapy so often poorly tolerated by infants and children. The main goal of this study is to assess, in a large series of patients, the prognosis of pediatric mediastinitis after our initial technique was applied with modifications.
Accepted for publication March 21, 2007. Address correspondence to Dr Durandy, Institut Hospitalier Jacques Cartier, Avenue du Noyer Lambert, Massy, 91300, France; e-mail: [email protected].
© 2007 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2007;84:423– 8) © 2007 by The Society of Thoracic Surgeons
Approval to conduct this retrospective anonymous study was obtained from our institutional review board, and the need for individual consent was waived. Sixty-four patients were treated for mediastinitis with closed chest drainage using Redon catheters, essentially 8F polyvinyl chloride tubing that connects to a lightweight plastic bottle so that patient mobility is retained. Mediastinitis was defined as a deep wound infection with frequent macroscopic sternal osteomyelitis. Whenever possible, and especially when there was purulent discharge from the wound, a fluid sample was collected for immediate bacteriologic examination and Gram stain. Usually a blood culture was obtained before pharmacologic treatment was started. Preoperatively, an intravenous antibiotic treatment based on empiric probability was always prescribed. As the incidence of methicillin-resistant staphylococcus was 35% in our unit, we routinely used a combination of vancomycin and gentamicin. Whenever a gram-negative bacillus was suspected, as in open-chest patients, a third-generation cephalosporin was added. This chemotherapy was secondarily adapted to the germ sensitivity. To prevent problems encountered with peripheral perfusion, a double-lumen venous catheter was inserted in the internal jugular or femoral vein. 0003-4975/07/$32.00 doi:10.1016/j.athoracsur.2007.03.064
GENERAL THORACIC
Postoperative Mediastinitis in Children: Improvement of Simple Primary Closed Drainage
GENERAL THORACIC
424
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
Ann Thorac Surg 2007;84:423– 8
Table 1. Isolated Mediastinitis in Neonates: 9 Patients Age (days)
Diagnosis
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
13 7 12 20 4 6 22 28 17
PA IVS Hypo AoA Hypo AoA VSD Hypo AoA TGA Ao S TPAVR Ao P W
Outflow enlargement Aortic arch repair Aortic arch repair VSD closure patch Aortic arch repair Arterial switch Valved conduit Confluence to left atrium anastomosis Defect closure patch
30 7 7 30 29 11 6 21 11
Pseu aer Stap epi Stap epi Stap au Stap au Stap au Coli Stap au Stap epi
22 12 12 14 15 13 18 12 17
Ao P W ⫽ aortopulmonary window; Ao S ⫽ aortic stenosis; Coli ⫽ Escherichia coli; Hypo AoA ⫽ hypoplasia of aortic arch; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; PA IVS ⫽ pulmonary atresia with intact ventricular septum; Pseu aer ⫽ Pseudomonas aeruginosa; Stap au ⫽ Staphylococcus aureus; Stap epi ⫽ Staphylococcus epidermitis; TGA ⫽ transposition of the great arteries; TPA VR ⫽ total pulmonary anomalous venous return; VSD ⫽ ventricular septal defect.
All patients underwent urgent surgical treatment. The sternum was totally reopened and two samples of retrosternal fluids were systematically collected for bacteriologic analysis. Debridement of all infected and devitalized tissue and local irrigation with povidone iodine (Betadine) was done. Sternal edges were revised, with a curette, until normal bleeding was present. Two to five Redon catheters were placed to drain fluids and collapse any mediastinal residual cavity. The sternal halves were approximated, and sternal closure was always performed using an absorbable suture (Vicryl, Ethicon, Johnson and
Johnson Intl, Somerville, NJ). Running absorbable suture was used to close the skin. Postoperative ventilatory support was either not necessary or limited to a few hours. Short-term ventilation decreased time spent in the intensive care unit, and most patients became ambulatory on the first postoperative day. The drainage from each Redon catheter was cultured daily. The Redon catheters were kept in place until fluids were sterile for 2 consecutive days, and then they were progressively withdrawn (2 cm every day). When the side holes of the drainage catheter were outside the chest wall, the vac-
Table 2. Isolated Mediastinitis in Infants: 20 Patients Age (months) 3 3 3 20 4 5 2 8 3 4 4 2 11 4 2 3 2 3 3 1
Diagnosis
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
PA IVS AVSD, Redo Ao S VSDs ASD AVSD ToF PS ⫹ Sup Ao S AVSD VSD AVSD Redo Scimitar syndrome AVSD Redux TGA, VSD TGA PS PA VSD Ao S, MI AVSD PA VSD VSD
Outflow enlargement Mitral valve repair Valvulotomy VSD closure patch ASD closure patch Total correction Total correction Pulmonary and aortic enlargement Total correction VSD closure patch Mitral valve repair Total correction with atrial patch Mitral valve replacement, pacemaker Arterial switch, VSD closure patch Pulmonary artery enlargement central shunt Repair with valved conduit Mitral valve replacement Total correction Outflow enlargement VSD closure patch
9 8 6 14 21 11 19 5 15 8 4 6 20 20 27 15 14 20 8 10
Stap epi Stap au Stap au Stap au Stap au Stap au Stap au Strepto G Stap au Stap au Stap au Stap au Stap au Stap au Stap au Stap epi Stap au Stap au Stap au Stap au
19 13 16 12 10 12 12 10 13 13 13 14 30 13 13 16 15 14 18 14
Ao S ⫽ aortic stenosis; ASD ⫽ atrial septal defect; AVSD ⫽ atrioventricular septal defect; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; MI ⫽ mitral valve incompetence; PA IVS ⫽ pulmonary atresia with intact ventricular septum; PA VSD ⫽ pulmonary atresia with ventricular septal defect; PS ⫽ pulmonary stenosis; Stap au ⫽ Staphylococcus aureus; Stap epi ⫽ Staphylococcus epidermitis; Strepto G ⫽ Streptococcus gordonii; Sup Ao S ⫽ supravalvar aortic stenosis; TGA ⫽ transposition of the great arteries; ToF ⫽ tetralogy of Fallot; VSD ⫽ ventricular septal defect; VSDs ⫽ multiple ventricular septal defects.
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
425
Table 3. Isolated Mediastinitis in Children: 11 Patients Age (years) 14 7 2 4, 5 13 8 3 8 3 5, 5 6
Diagnosis
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
Ao S SV AVSD Redo Ao S PA VSD SV ASD Cor TGA AVSD, MS TGA, PS Shone
Konno Fontan Mitral valve replacement, pacemaker Ross Total correction Fontan ASD closure patch Switch ⫹ Senning Mitral valve replacement REV procedure Ross
10 4 13 12 20 13 10 11 20 11 7
Stap au Stap au Stap au Stap epi Stap au Stap au Stap au Stap au Stap au Stap au Stap au
12 12 17 13 12 13 16 30 13 14 15
Ao S ⫽ aortic stenosis; ASD ⫽ atrial septal defect; AVSD ⫽ atrioventricular septal defect; Cor TGA ⫽ corrected transposition of the great arteries; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; MS ⫽ mitral valve stenosis; PA VSD ⫽ pulmonary atresia with ventricular septal defect; PS ⫽ pulmonary stenosis; REV ⫽ translocation of the pulmonary artery onto the right ventricle; Shone ⫽ Shone syndrome (supravalvar mitral ring, parachute mitral valve, subvalvar aortic obstruction, bicuspid aortic valve, aortic coarctation); Stap au ⫽ Staphylococcus aureus; Stap epi ⫽ Staphylococcus epidermitis; SV ⫽ single ventricle; TGA ⫽ transposition of the great arteries.
uum within the drainage unit was lost and the catheter was removed. The venous catheter was maintained 24 hours after removal of the last Redon catheter. Patients were discharged soon after venous line withdrawal without any treatment. None of the patients were treated as outpatients. In rare cases of delay in mediastinal sterilization, confirmed by culture of Redon catheter fluid, antibiotic dosage was performed in the blood and the mediastinal fluid to assess antibiotic diffusion in the infected area. A third antibiotic or an increased dosage of ongoing therapy was prescribed. The overall follow-up period of this series varied from 7 months to 12 years. The data collected included age, diagnosis, surgical treatment, the percentage of delayed sternal closure after initial surgical cure, the delay between initial surgery and mediastinal infection, infective species, the result of blood culture, the time to extubation, the time to when the effluent culture was negative for microorganisms, the duration of intravenous antibiotic course, and the duration of hospital stay. To analyze the outcomes of our treatment and the prognosis of mediastinitis, the patients were classified
into three groups: group 1—patients with isolated mediastinitis; group 2—patients with mediastinitis associated with endocarditis; and group 3—patients with mediastinitis associated with other major organ failure.
Results From January 1994 to December 2005, 3,650 pediatric patients underwent median sternotomy. During this period, mediastinitis occurred in 64 patients (1.7%). In every case, diagnosis was confirmed by the positive culture of a perioperative retrosternal bacteriologic sample. The 64 patients were classified into three groups. Data collected from each group are summarized in Tables 1 to 5. Time to mediastinal sterilization according to germ type is indicated in Table 6. The first group with isolated mediastinitis comprised 40 patients: 9 neonates, with a median patient age of 13 days (range, 4 to 28 days); 20 infants, with a median patient age of 3 months (range, 1 to 20 months); and 11 children, with a median patient age of 6 years (range, 2 to 14 years). In this group there were no complications, and ventilatory time was short. The median length of hospital
Table 4. Mediastinitis Associated With Endocarditis: 7 Patients Age 4 mo 3y 5 days 8 days 4y 2y 3 mo
Diagnosis
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
VSD PA VSD PA IVS TGA AVSD VSD VSD
VSD closure patch Repair with valved conduit Outflow enlargement Arterial switch ASD closure patch VSD closure patch VSD closure patch
12 5 9 5 16 26 5
Stap au Pneumo Stap au Stap au Stap au Stap au Stap au
32 20 (death) 54 36 45 40 37
ASD ⫽ atrial septal defect; AVSD ⫽ atrio ventricular septal defect; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; PA IVS ⫽ pulmonary atresia with intact ventricular septum; PA VSD ⫽ pulmonary atresia with ventricular septal defect; Pneumo ⫽ Pneumococcus; Stap au ⫽ Staphylococcus aureus; TGA ⫽ transposition of the great arteries; VSD ⫽ ventricular septal defect.
GENERAL THORACIC
Ann Thorac Surg 2007;84:423– 8
GENERAL THORACIC
426
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
Ann Thorac Surg 2007;84:423– 8
Table 5. Mediastinitis Associated With Organ Failure: 17 Patients Age
Diagnosis
2 mo 8 days 2 mo 1 days
AVSD Shone, redo Shone, redo TPAVR
2 mo 11 days 8 days 1 mo 8y 2 mo 4 mo 3 mo 1 mo 6 mo 1y 9 mo 15 y
TGA, redo Ao S Hypo Ao A TGA, Hypo Ao A TGA, redo Ao S Sub Ao S PA VSD AVSD ToF Cor TGA PA VSD TGA, PS
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
Repair Aortoplasty Aortoplasty Confluence to left atrium anastomosis Coronaroplasty Valvulotomy Aortic arch repair Arterial switch, aortoplasty Coronaroplasty Ross Outflow enlargement Repair with valved conduit Total correction Total correction Switch ⫹ Senning Repair with valved conduit REV procedure
35 12 30 25
Stap au Pseu aer Pseu aer Stap epi
19 23 17 18
7 5 5 16 8 5 7 11 15 3 16 10 37
Stap epi, Coli Stap au Coli Stap au Stap au Stap epi Stap au, Coli Stap epi Entero Stap au, Coli Stap au Stap au Stap au
10 death 18 21 20 18 32 26 18 30 15 75 death 28 36
Ao S ⫽ aortic stenosis; AVSD ⫽ atrioventricular septal defect; Coli ⫽ Escherichia coli; Cor TGA ⫽ corrected transposition of the great arteries; Enter ⫽ Enterobacter; Hypo Ao A ⫽ hypoplasia of aortic arch; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; PA VSD ⫽ pulmonary atresia with ventricular septal defect; PS ⫽ pulmonary stenosis; Pseu aer ⫽ Pseudomonas aeruginosa; REV ⫽ translocation of the pulmonary artery onto the right ventricle; Shone ⫽ Shone syndrome (supravalvar mitral ring, parachute mitral valve, subvalvar aortic obstruction, bicuspid aortic valve, aortic coarctation); Stap au ⫽ Staphylococcus aureus; Stap epi ⫽ Staphylococcus epidermitis; Sub Ao S ⫽ subvalvar aortic stenosis; TGA ⫽ transposition of the great arteries; ToF ⫽ tetralogy of Fallot; TPAVR ⫽ total pulmonary anomalous venous return.
stay was 13 days (range, 10 to 30 days). The overall eradication rate was 100%. Less than 2 weeks of intravenous therapy was sufficient for mediastinitis sterilization (median length, 11 days; range, 7 to 28 days) and recovery, without any recurrence of infection. Seven patients (11%) were in the second group with mediastinitis and endocarditis. In every case, endocarditis suspected from clinical and bacteriologic features was confirmed by echocardiographic data. The duration of intravenous antibiotic therapy was related to the heart infection. The delay for mediastinal sterilization was 8 days, and the eradication rate was also 100%. One patient with Pneumococcus endocarditis was reoperated on for removal of a valved conduit with false aneurysm. Perioperative mediastinal bacteriologic samples were negative. The patient died in the operating room of hemorrhagic and hemodynamic complications. The third group comprised 17 patients (27%) with mediastinitis and other organ failure. The main organ involved was the lung in 9 patients with acute respiratory distress syndrome, the kidney in 6 patients with acute renal failure (2 of them required peritoneal dialysis), and the heart and lung in 2 patients treated with extracorporeal support. In this group, the hemodynamic instability prefaces the mediastinitis, as attested by the high incidence of open-chest patients (72%). Renal and respiratory failure were always present before sepsis and diagnosis of mediastinitis. The duration of mechanical ventilation is linked with the high incidence of acute respiratory distress syndrome (53% of patients). The median length of stay was 20 days (range, 18 to 36 days).
In this more critically ill group of patients there was also 100% eradication of mediastinal sepsis. The median length of intravenous antibiotic course was 15 days (range, 10 to 31 days). Two patients died with a sterile mediastinal drainage. One 2 month-old boy had a neonatal arterial switch operation followed within a week by reoperation for pulmonary stenosis. He was, 2 months later, twice reoperated on for right coronary artery stenosis and persistent pulmonary stenosis. He finally exhibited mediastinitis and died 10 days later of acute respiratory distress syndrome, but had a germ-free mediastinal drainage. The second patient had a double switch procedure (arterial switch and Senning) with postoperative complications treated with open chest and two periods of extracorporeal membrane oxygenation lasting 15 days and 48 days. The patient died 48 hours after weaning from extracorporeal membrane oxygenation with neurologic complication. At that time the mediastinum was germ free. Table 6. Delay of Mediastinal Sterilization With Staphylococcus or Gram-Negative Bacterium Organism Staphylococcus aureus Staphylococcus epidermitis Staphylococcus, methicillin sensitive Staphylococcus, methicillin resistant Gram-negative bacilli
Delay of Sterilization (days) Median and Range 5 (1–10) 4 (1–13) 4 (1–13) 4 (1–10) 8 (3–9)
Three patients, one in each group, experienced delayed sterilization of mediastinal fluid with a rapid response to increased chemotherapy. All of them, including one case with reduced susceptibility to vancomycin strain of staphylococcus [13], were rapidly cured by prolongation of drainage for a few days and alteration of initial chemotherapy.
Comment The patients from this series are similar to those described in the literature. The incidence of postoperative mediastinitis in children varies between 0.2% and 5% [4, 14, 15]. The incidence in our group (1.7%) was high but within the most generally accepted rate of 1% to 2%. Delayed sternal closure is a classic risk factor [15–17]. In the 3,650 patients operated on during this study, the overall percentage of delayed sternal closure is 5%, compared with 22% in group 1, 28% in group 2, and 72% in group 3. As stated by others [18, 19], the incidence of blood culture positive for microorganisms is far above that of endocarditis. Our incidence of preoperative blood culture positive for microorganisms (61%) was probably underestimated, because that was assessed in only 57 of the 64 patients. Staphylococcus was the main causative organism, accounting for 83% of those cultured (staphylococcus was present in 3 patients with two species isolated from the mediastinum), and gram-negative bacilli were mainly present when sternal closure was delayed. These data are also in concordance with previous studies [17, 18]. Our technique and results compare favorably with that of others. The open-chest technique with irrigation and drainage is a long, often painful technique with frequent dressing. Mechanical ventilation is prescribed, which has its own inherent risks, and is not easily tolerated by pediatric patients. Vacuum-assisted closure therapy is a delayed closure technique with dressing changes, performed most often on alternate days [20]. There is little information about skin tolerance of watertight dressing in infected infants, the need for sedation and mechanical ventilation during therapy (or at least during dressing changes), and the correct level of negative pressure and choice of continuous or intermittent pressure therapy. Moreover, this technique prescribes confinement in bed for days or weeks. Seven cases of cardiac rupture have been described during topical negative-pressure therapy in adults [21, 22]. Muscle flap is associated with delayed or primary closure. However, there are doubts about the functional and cosmetic results of major muscle flaps, with risk of breast hypoplasia and sensory loss occurring in adolescent girls [5]. The goal of muscle flap is to fill the sternal defect with well-vascularized healthy tissue. We believe that strong negative pressure after meticulous debridement achieves the same effect with a less aggressive approach. Finally, primary closure with short-term drainage advocated by the Ann Arbor group [8] is not very different
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
427
Table 7. Duration of Intravenous Antibiotic Therapy in Recent Studies of Mediastinitis in Pediatric Patients
References Sung et al [5] Ohye et al [8] Tortoriello et al [14] Al-sehly et al [16] a
Number of Patients
Duration of IV Antibiotic Therapy
Hospital Stay (days) Median and Range
11 57 15 29
4–6 weeksa 6 weeks 4–6 weeks 4–8 weeks
24 (8–140) 12 (4–88) 42.5 (16–163) N/A
6 weeks in patients with positive blood culture.
IV ⫽ intravenous;
N/A ⫽ not available.
from our policy. In their series, 3 of 42 patients (7%) were reoperated on for continuing sepsis. In our experience, the daily culture of mediastinal effluent has been efficient in preventing reoperation. This advantage is illustrated in the 3 patients with delayed sterilization followed by the rapid negative status of fluid culture after therapeutic modification. Our technique is associated with a short course of intravenous antibiotics, about 2 weeks versus 4 to 8 weeks most often recommended in the literature (Table 7). We have learned from our technique that mediastinal fluid remains infected a few days after surgical revision. The good results obtained with short-term chemotherapy are probably related to the efficiency of prolonging the drainage until the mediastinal effluent cultures negative for microorganisms. The outcome is a decreased hospital stay (a major factor in decreasing restraint owing to mediastinitis). In contrast to adult surgery, the prognosis of mediastinitis in pediatric surgery is much better with an overall success rate of approximately 100% [4, 5, 8, 14]. In most cases, death is not directly related to uncontrolled mediastinal sepsis [8, 16]. In this series, mortality was related to a patient’s associated disease and was not related to mediastinitis or therapeutic complications. Pneumococcus is uncommon in postoperative mediastinitis but associated with poor prognosis [23, 24]. In our patient, an extreme fragility of the inflammatory tissues was considered as the main cause of perioperative hemorrhage and death. The two other patients had well-documented nonseptic cause of death. Through a large series of patients we have demonstrated that primary closure with Redon catheter drainage is always efficient in the eradication of infection. Therefore, mediastinitis is not a direct cause of death in our experience. The modification of our initial therapy toward shorter drainage time and a shorter course of antibiotic therapy did not impair the good results obtained with this therapy.
References 1. Shumacker HB, Mandelbaum I. Continuous antibiotic irrigation in the treatment of infection. Arch Surg 1963;86:384 –7.
GENERAL THORACIC
Ann Thorac Surg 2007;84:423– 8
GENERAL THORACIC
428
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
2. Ohuchi S, Kawazoe K, Ishihara K, Izumoto H, Eishi K. Management with closed irrigation for post-sternotomy mediastinitis: experience with the use of electrolyzed strong acid aqueous solution. Jpn J Thorac Cardiovasc Surg 2003; 51:511– 4. 3. Stiegel RM, Beasley ME, Sink JD, et al. Management of postoperative mediastinitis in infants and children by muscle flap rotation. Ann Thorac Surg 1988;46:45– 6. 4. Erez E, Katz M, Sharoni E, et al. Pectoralis major muscle flap for deep sternal wound infection in neonates. Ann Thorac Surg 2000;69:572–7. 5. Sung K, Jun TG, Park PW, Park KH, Lee YT, Yang JH. Management of deep sternal infection in infants and children with advanced pectoralis major muscle flaps. Ann Thorac Surg 2004;77:1371–5. 6. Agarwal JP, Ogilvie M, Wu LC, et al. Vacuum-assisted closure for sternal wounds: a first-line therapeutic management approach. Plast Reconstr Surg 2005;116:1035– 43. 7. Ramnarine IR, McLean A, Pollock JCS. Vacuum-assisted closure in the pediatric patient with post-cardiotomy mediastinitis. Eur J Cardiothorac Surg 2002;22:1029 –31. 8. Ohye RG, Maniker RB, Graves HL, Devaney EJ, Bove EL. Primary closure for postoperative mediastinitis in children. J Thorac Cardiovasc Surg 2004;128:480 – 6. 9. Durandy Y, Batisse A, Bourel P, Dibie A, Lemoine G, Lecompte Y. Mediastinal infection after cardiac operation: a simple closed technique. J Thorac Cardiovasc Surg 1989;97:282–5. 10. Calvat S, Trouillet JL, Nataf P, Vuagnat A, Chastre J, Gibert C. Closed drainage using Redon catheters for local treatment of poststernotomy mediastinitis. Ann Thorac Surg 1996;61: 195–201. 11. Kirsh M, Mekontso-Dessap A, Houel R, Giroud E, Hillion ML, Loisance DY. Closed drainage using Redon catheters for poststernotomy mediastinitis: results and risk factors for adverse outcome. Ann Thorac Surg 2001;71:1580 – 6. 12. Berg HF, Brands WG, Van Geldorp TR, KluytmansVandenBergh FQ, Kluytmans JA. Comparison between closed drainage techniques for the treatment of postoperative mediastinitis. Ann Thorac Surg 2000;70:924 –9. 13. Hulin S, Durandy Y. Postoperative mediastinitis due to methicillin-resistant Staphylococcus epidermidis with low sensitivity to vancomycin. Arch Pediatr 1997;41:1204 – 6.
Ann Thorac Surg 2007;84:423– 8
14. Tortoriello TA, Friedman JD, McKenzie ED, et al. Mediastinitis after pediatric cardiac surgery: a 15-year experience at a single institution. Ann Thorac Surg 2003;76:1655– 60. 15. Mehta PA, Cunningham CK, Colella CB, Alferis G, Weiner LB. Risk factors for sternal wound and other infections in pediatric cardiac surgery patients. Pediatr Infect Dis J 2000; 19:1000 – 4. 16. Al-sehly AA, Robinson JL, Lee BE, et al. Pediatric poststernotomy mediastinitis. Ann Thorac Surg 2005;80:2314 –20. 17. Long CB, Shah SS, Lautenbach E, et al. Postoperative mediastinitis in children: epidemiology, microbiology and risk factors for Gram-negative pathogens. Pediatr Infect Dis J 2005;24:315–9. 18. San Juan R, Aguado JM, Lopez MJ, et al. Accuracy of blood culture for early diagnosis of mediastinitis in febrile patients after cardiac surgery. Eur. J Clin Microbiol Infect Dis 2005; 24:182–9. 19. Fowler VG, Kaye KS, Simel DL, et al. Staphylococcus aureus bacteremia after median sternotomy: clinical utility of blood culture results in the identification of postoperative mediastinitis. Circulation 2003;108:73– 8. 20. Loos B, Kopp J, Kneser U, Weyand M, Horch RE. The importance of vacuum therapy in the treatment of sternal osteomyelitis from the plastic surgeons point of view. Zentralbl Chir 2006;131:124 – 8. 21. Sartipy U, Lockowandt U, Gäbel J, Jidéus L, Dellgren G. Cardiac rupture during vacuum-assisted closure therapy. Ann Thorac Surg 2006;82:1110 –1. 22. Abu-Omar Y, Naik MJ, Catarino PA, Ratnatunga C. Right ventricular rupture during use of high-pressure suction drainage in the management of poststernotomy mediastinitis [letter]. Ann Thorac Surg 2003;76:974. 23. Souza Neto EP, Celard M, Durand PG, Ninet J, Lehot JJ. Fulminant mediastinitis from Streptococcus pneumoniae following cardiac surgery. Ann Fr Anesth Reanim 2002;21: 603–5. 24. Maroto LC, Aguado JM, Carrascal Y, Rodriguez JE, Cortina JM, Rufilanchas JJ. Fulminant mediastinitis due to Streptococcus pneumoniae following cardiac surgery. Clin Inf Dis 1996;22:594 –5.
Background. Mediastinitis is a significant cause of postoperative morbidity. In 1989, we proposed simple primary closed drainage as a new treatment. Our goal is to describe improvements made to the original technique. Methods. After wound debridement, infected areas were drained with Redon catheters connected to strong negative-pressure drainage bottles. Mediastinal effluents were cultured every day, and the catheters were withdrawn when the effluent culture was negative for microorganisms. Patients were classified into three groups: isolated mediastinitis (group 1), mediastinitis associated with endocarditis (group 2), and mediastinitis associated with other organ failure (group 3). Results. Sixty-four patients were treated during a 10year period: 15 neonates, 33 infants, and 16 children. Group 1 consisted of 40 patients. The time to mediastinal sterilization was 4 days (range, 1 to 14 days), and the antibiotic course was 11 days (range, 7 to 28 days), with a hospital stay of 13 days (range, 10 to 30 days). No deaths
occurred in this group. Group 2 consisted of 7 patients. The time to mediastinal sterilization was 8 days (range, 3 to 10 days), and the antibiotic course was 30 days (range, 26 to 37 days), with a hospital stay of 37 days (range, 20 to 54 days). One patient in group 2 did not survive. Group 3 consisted of 17 patients. The time to mediastinal sterilization was 6 days (range, 1 to 10 days), and the antibiotic course was 15 days (range, 10 to 31 days), with a hospital stay of 20 days (range, 18 to 36 days). Two patients in group 3 did not survive. None of the deaths was directly related to mediastinitis, as the mediastinum was sterile in all 3 patients before death. Conclusions. This simple treatment was efficient and reliable in achieving mediastinal sterilization. In addition, short antibiotic courses decreased restraint, which is poorly tolerated in pediatric patients.
M
Material and Methods
ediastinitis is a major complication ensuing pediatric cardiac surgery. Its treatment remains controversial. The first part of therapeutic management is always surgical revision, which is completed with one of the following procedures: closed irrigation, an old technique first described in 1963 [1] and still used [2]; delayed or primary muscle flaps [3–5]; vacuum-assisted closure, first described in adults but also used in infants [6, 7]; or primary closure with short-term drainage [8]. In 1989 we proposed simple primary closure with drainage through a Redon catheter [9], which was subsequently practiced by others [10 –12]. Since then we have learned from much experience that it is possible to shorten the drainage time and course of chemotherapy. This strategy also demonstrates benefits in terms of decreasing the physical and psychological effects of hospital stays and intravenous therapy so often poorly tolerated by infants and children. The main goal of this study is to assess, in a large series of patients, the prognosis of pediatric mediastinitis after our initial technique was applied with modifications.
Accepted for publication March 21, 2007. Address correspondence to Dr Durandy, Institut Hospitalier Jacques Cartier, Avenue du Noyer Lambert, Massy, 91300, France; e-mail: [email protected].
© 2007 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2007;84:423– 8) © 2007 by The Society of Thoracic Surgeons
Approval to conduct this retrospective anonymous study was obtained from our institutional review board, and the need for individual consent was waived. Sixty-four patients were treated for mediastinitis with closed chest drainage using Redon catheters, essentially 8F polyvinyl chloride tubing that connects to a lightweight plastic bottle so that patient mobility is retained. Mediastinitis was defined as a deep wound infection with frequent macroscopic sternal osteomyelitis. Whenever possible, and especially when there was purulent discharge from the wound, a fluid sample was collected for immediate bacteriologic examination and Gram stain. Usually a blood culture was obtained before pharmacologic treatment was started. Preoperatively, an intravenous antibiotic treatment based on empiric probability was always prescribed. As the incidence of methicillin-resistant staphylococcus was 35% in our unit, we routinely used a combination of vancomycin and gentamicin. Whenever a gram-negative bacillus was suspected, as in open-chest patients, a third-generation cephalosporin was added. This chemotherapy was secondarily adapted to the germ sensitivity. To prevent problems encountered with peripheral perfusion, a double-lumen venous catheter was inserted in the internal jugular or femoral vein. 0003-4975/07/$32.00 doi:10.1016/j.athoracsur.2007.03.064
GENERAL THORACIC
Postoperative Mediastinitis in Children: Improvement of Simple Primary Closed Drainage
GENERAL THORACIC
424
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
Ann Thorac Surg 2007;84:423– 8
Table 1. Isolated Mediastinitis in Neonates: 9 Patients Age (days)
Diagnosis
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
13 7 12 20 4 6 22 28 17
PA IVS Hypo AoA Hypo AoA VSD Hypo AoA TGA Ao S TPAVR Ao P W
Outflow enlargement Aortic arch repair Aortic arch repair VSD closure patch Aortic arch repair Arterial switch Valved conduit Confluence to left atrium anastomosis Defect closure patch
30 7 7 30 29 11 6 21 11
Pseu aer Stap epi Stap epi Stap au Stap au Stap au Coli Stap au Stap epi
22 12 12 14 15 13 18 12 17
Ao P W ⫽ aortopulmonary window; Ao S ⫽ aortic stenosis; Coli ⫽ Escherichia coli; Hypo AoA ⫽ hypoplasia of aortic arch; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; PA IVS ⫽ pulmonary atresia with intact ventricular septum; Pseu aer ⫽ Pseudomonas aeruginosa; Stap au ⫽ Staphylococcus aureus; Stap epi ⫽ Staphylococcus epidermitis; TGA ⫽ transposition of the great arteries; TPA VR ⫽ total pulmonary anomalous venous return; VSD ⫽ ventricular septal defect.
All patients underwent urgent surgical treatment. The sternum was totally reopened and two samples of retrosternal fluids were systematically collected for bacteriologic analysis. Debridement of all infected and devitalized tissue and local irrigation with povidone iodine (Betadine) was done. Sternal edges were revised, with a curette, until normal bleeding was present. Two to five Redon catheters were placed to drain fluids and collapse any mediastinal residual cavity. The sternal halves were approximated, and sternal closure was always performed using an absorbable suture (Vicryl, Ethicon, Johnson and
Johnson Intl, Somerville, NJ). Running absorbable suture was used to close the skin. Postoperative ventilatory support was either not necessary or limited to a few hours. Short-term ventilation decreased time spent in the intensive care unit, and most patients became ambulatory on the first postoperative day. The drainage from each Redon catheter was cultured daily. The Redon catheters were kept in place until fluids were sterile for 2 consecutive days, and then they were progressively withdrawn (2 cm every day). When the side holes of the drainage catheter were outside the chest wall, the vac-
Table 2. Isolated Mediastinitis in Infants: 20 Patients Age (months) 3 3 3 20 4 5 2 8 3 4 4 2 11 4 2 3 2 3 3 1
Diagnosis
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
PA IVS AVSD, Redo Ao S VSDs ASD AVSD ToF PS ⫹ Sup Ao S AVSD VSD AVSD Redo Scimitar syndrome AVSD Redux TGA, VSD TGA PS PA VSD Ao S, MI AVSD PA VSD VSD
Outflow enlargement Mitral valve repair Valvulotomy VSD closure patch ASD closure patch Total correction Total correction Pulmonary and aortic enlargement Total correction VSD closure patch Mitral valve repair Total correction with atrial patch Mitral valve replacement, pacemaker Arterial switch, VSD closure patch Pulmonary artery enlargement central shunt Repair with valved conduit Mitral valve replacement Total correction Outflow enlargement VSD closure patch
9 8 6 14 21 11 19 5 15 8 4 6 20 20 27 15 14 20 8 10
Stap epi Stap au Stap au Stap au Stap au Stap au Stap au Strepto G Stap au Stap au Stap au Stap au Stap au Stap au Stap au Stap epi Stap au Stap au Stap au Stap au
19 13 16 12 10 12 12 10 13 13 13 14 30 13 13 16 15 14 18 14
Ao S ⫽ aortic stenosis; ASD ⫽ atrial septal defect; AVSD ⫽ atrioventricular septal defect; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; MI ⫽ mitral valve incompetence; PA IVS ⫽ pulmonary atresia with intact ventricular septum; PA VSD ⫽ pulmonary atresia with ventricular septal defect; PS ⫽ pulmonary stenosis; Stap au ⫽ Staphylococcus aureus; Stap epi ⫽ Staphylococcus epidermitis; Strepto G ⫽ Streptococcus gordonii; Sup Ao S ⫽ supravalvar aortic stenosis; TGA ⫽ transposition of the great arteries; ToF ⫽ tetralogy of Fallot; VSD ⫽ ventricular septal defect; VSDs ⫽ multiple ventricular septal defects.
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
425
Table 3. Isolated Mediastinitis in Children: 11 Patients Age (years) 14 7 2 4, 5 13 8 3 8 3 5, 5 6
Diagnosis
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
Ao S SV AVSD Redo Ao S PA VSD SV ASD Cor TGA AVSD, MS TGA, PS Shone
Konno Fontan Mitral valve replacement, pacemaker Ross Total correction Fontan ASD closure patch Switch ⫹ Senning Mitral valve replacement REV procedure Ross
10 4 13 12 20 13 10 11 20 11 7
Stap au Stap au Stap au Stap epi Stap au Stap au Stap au Stap au Stap au Stap au Stap au
12 12 17 13 12 13 16 30 13 14 15
Ao S ⫽ aortic stenosis; ASD ⫽ atrial septal defect; AVSD ⫽ atrioventricular septal defect; Cor TGA ⫽ corrected transposition of the great arteries; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; MS ⫽ mitral valve stenosis; PA VSD ⫽ pulmonary atresia with ventricular septal defect; PS ⫽ pulmonary stenosis; REV ⫽ translocation of the pulmonary artery onto the right ventricle; Shone ⫽ Shone syndrome (supravalvar mitral ring, parachute mitral valve, subvalvar aortic obstruction, bicuspid aortic valve, aortic coarctation); Stap au ⫽ Staphylococcus aureus; Stap epi ⫽ Staphylococcus epidermitis; SV ⫽ single ventricle; TGA ⫽ transposition of the great arteries.
uum within the drainage unit was lost and the catheter was removed. The venous catheter was maintained 24 hours after removal of the last Redon catheter. Patients were discharged soon after venous line withdrawal without any treatment. None of the patients were treated as outpatients. In rare cases of delay in mediastinal sterilization, confirmed by culture of Redon catheter fluid, antibiotic dosage was performed in the blood and the mediastinal fluid to assess antibiotic diffusion in the infected area. A third antibiotic or an increased dosage of ongoing therapy was prescribed. The overall follow-up period of this series varied from 7 months to 12 years. The data collected included age, diagnosis, surgical treatment, the percentage of delayed sternal closure after initial surgical cure, the delay between initial surgery and mediastinal infection, infective species, the result of blood culture, the time to extubation, the time to when the effluent culture was negative for microorganisms, the duration of intravenous antibiotic course, and the duration of hospital stay. To analyze the outcomes of our treatment and the prognosis of mediastinitis, the patients were classified
into three groups: group 1—patients with isolated mediastinitis; group 2—patients with mediastinitis associated with endocarditis; and group 3—patients with mediastinitis associated with other major organ failure.
Results From January 1994 to December 2005, 3,650 pediatric patients underwent median sternotomy. During this period, mediastinitis occurred in 64 patients (1.7%). In every case, diagnosis was confirmed by the positive culture of a perioperative retrosternal bacteriologic sample. The 64 patients were classified into three groups. Data collected from each group are summarized in Tables 1 to 5. Time to mediastinal sterilization according to germ type is indicated in Table 6. The first group with isolated mediastinitis comprised 40 patients: 9 neonates, with a median patient age of 13 days (range, 4 to 28 days); 20 infants, with a median patient age of 3 months (range, 1 to 20 months); and 11 children, with a median patient age of 6 years (range, 2 to 14 years). In this group there were no complications, and ventilatory time was short. The median length of hospital
Table 4. Mediastinitis Associated With Endocarditis: 7 Patients Age 4 mo 3y 5 days 8 days 4y 2y 3 mo
Diagnosis
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
VSD PA VSD PA IVS TGA AVSD VSD VSD
VSD closure patch Repair with valved conduit Outflow enlargement Arterial switch ASD closure patch VSD closure patch VSD closure patch
12 5 9 5 16 26 5
Stap au Pneumo Stap au Stap au Stap au Stap au Stap au
32 20 (death) 54 36 45 40 37
ASD ⫽ atrial septal defect; AVSD ⫽ atrio ventricular septal defect; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; PA IVS ⫽ pulmonary atresia with intact ventricular septum; PA VSD ⫽ pulmonary atresia with ventricular septal defect; Pneumo ⫽ Pneumococcus; Stap au ⫽ Staphylococcus aureus; TGA ⫽ transposition of the great arteries; VSD ⫽ ventricular septal defect.
GENERAL THORACIC
Ann Thorac Surg 2007;84:423– 8
GENERAL THORACIC
426
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
Ann Thorac Surg 2007;84:423– 8
Table 5. Mediastinitis Associated With Organ Failure: 17 Patients Age
Diagnosis
2 mo 8 days 2 mo 1 days
AVSD Shone, redo Shone, redo TPAVR
2 mo 11 days 8 days 1 mo 8y 2 mo 4 mo 3 mo 1 mo 6 mo 1y 9 mo 15 y
TGA, redo Ao S Hypo Ao A TGA, Hypo Ao A TGA, redo Ao S Sub Ao S PA VSD AVSD ToF Cor TGA PA VSD TGA, PS
Surgical Treatment
Interval S/M (days)
Organism
Interval M/HD (days)
Repair Aortoplasty Aortoplasty Confluence to left atrium anastomosis Coronaroplasty Valvulotomy Aortic arch repair Arterial switch, aortoplasty Coronaroplasty Ross Outflow enlargement Repair with valved conduit Total correction Total correction Switch ⫹ Senning Repair with valved conduit REV procedure
35 12 30 25
Stap au Pseu aer Pseu aer Stap epi
19 23 17 18
7 5 5 16 8 5 7 11 15 3 16 10 37
Stap epi, Coli Stap au Coli Stap au Stap au Stap epi Stap au, Coli Stap epi Entero Stap au, Coli Stap au Stap au Stap au
10 death 18 21 20 18 32 26 18 30 15 75 death 28 36
Ao S ⫽ aortic stenosis; AVSD ⫽ atrioventricular septal defect; Coli ⫽ Escherichia coli; Cor TGA ⫽ corrected transposition of the great arteries; Enter ⫽ Enterobacter; Hypo Ao A ⫽ hypoplasia of aortic arch; Interval M/HD ⫽ interval between mediastinitis and hospital discharge; Interval S/M ⫽ interval between surgery and mediastinitis diagnosis; PA VSD ⫽ pulmonary atresia with ventricular septal defect; PS ⫽ pulmonary stenosis; Pseu aer ⫽ Pseudomonas aeruginosa; REV ⫽ translocation of the pulmonary artery onto the right ventricle; Shone ⫽ Shone syndrome (supravalvar mitral ring, parachute mitral valve, subvalvar aortic obstruction, bicuspid aortic valve, aortic coarctation); Stap au ⫽ Staphylococcus aureus; Stap epi ⫽ Staphylococcus epidermitis; Sub Ao S ⫽ subvalvar aortic stenosis; TGA ⫽ transposition of the great arteries; ToF ⫽ tetralogy of Fallot; TPAVR ⫽ total pulmonary anomalous venous return.
stay was 13 days (range, 10 to 30 days). The overall eradication rate was 100%. Less than 2 weeks of intravenous therapy was sufficient for mediastinitis sterilization (median length, 11 days; range, 7 to 28 days) and recovery, without any recurrence of infection. Seven patients (11%) were in the second group with mediastinitis and endocarditis. In every case, endocarditis suspected from clinical and bacteriologic features was confirmed by echocardiographic data. The duration of intravenous antibiotic therapy was related to the heart infection. The delay for mediastinal sterilization was 8 days, and the eradication rate was also 100%. One patient with Pneumococcus endocarditis was reoperated on for removal of a valved conduit with false aneurysm. Perioperative mediastinal bacteriologic samples were negative. The patient died in the operating room of hemorrhagic and hemodynamic complications. The third group comprised 17 patients (27%) with mediastinitis and other organ failure. The main organ involved was the lung in 9 patients with acute respiratory distress syndrome, the kidney in 6 patients with acute renal failure (2 of them required peritoneal dialysis), and the heart and lung in 2 patients treated with extracorporeal support. In this group, the hemodynamic instability prefaces the mediastinitis, as attested by the high incidence of open-chest patients (72%). Renal and respiratory failure were always present before sepsis and diagnosis of mediastinitis. The duration of mechanical ventilation is linked with the high incidence of acute respiratory distress syndrome (53% of patients). The median length of stay was 20 days (range, 18 to 36 days).
In this more critically ill group of patients there was also 100% eradication of mediastinal sepsis. The median length of intravenous antibiotic course was 15 days (range, 10 to 31 days). Two patients died with a sterile mediastinal drainage. One 2 month-old boy had a neonatal arterial switch operation followed within a week by reoperation for pulmonary stenosis. He was, 2 months later, twice reoperated on for right coronary artery stenosis and persistent pulmonary stenosis. He finally exhibited mediastinitis and died 10 days later of acute respiratory distress syndrome, but had a germ-free mediastinal drainage. The second patient had a double switch procedure (arterial switch and Senning) with postoperative complications treated with open chest and two periods of extracorporeal membrane oxygenation lasting 15 days and 48 days. The patient died 48 hours after weaning from extracorporeal membrane oxygenation with neurologic complication. At that time the mediastinum was germ free. Table 6. Delay of Mediastinal Sterilization With Staphylococcus or Gram-Negative Bacterium Organism Staphylococcus aureus Staphylococcus epidermitis Staphylococcus, methicillin sensitive Staphylococcus, methicillin resistant Gram-negative bacilli
Delay of Sterilization (days) Median and Range 5 (1–10) 4 (1–13) 4 (1–13) 4 (1–10) 8 (3–9)
Three patients, one in each group, experienced delayed sterilization of mediastinal fluid with a rapid response to increased chemotherapy. All of them, including one case with reduced susceptibility to vancomycin strain of staphylococcus [13], were rapidly cured by prolongation of drainage for a few days and alteration of initial chemotherapy.
Comment The patients from this series are similar to those described in the literature. The incidence of postoperative mediastinitis in children varies between 0.2% and 5% [4, 14, 15]. The incidence in our group (1.7%) was high but within the most generally accepted rate of 1% to 2%. Delayed sternal closure is a classic risk factor [15–17]. In the 3,650 patients operated on during this study, the overall percentage of delayed sternal closure is 5%, compared with 22% in group 1, 28% in group 2, and 72% in group 3. As stated by others [18, 19], the incidence of blood culture positive for microorganisms is far above that of endocarditis. Our incidence of preoperative blood culture positive for microorganisms (61%) was probably underestimated, because that was assessed in only 57 of the 64 patients. Staphylococcus was the main causative organism, accounting for 83% of those cultured (staphylococcus was present in 3 patients with two species isolated from the mediastinum), and gram-negative bacilli were mainly present when sternal closure was delayed. These data are also in concordance with previous studies [17, 18]. Our technique and results compare favorably with that of others. The open-chest technique with irrigation and drainage is a long, often painful technique with frequent dressing. Mechanical ventilation is prescribed, which has its own inherent risks, and is not easily tolerated by pediatric patients. Vacuum-assisted closure therapy is a delayed closure technique with dressing changes, performed most often on alternate days [20]. There is little information about skin tolerance of watertight dressing in infected infants, the need for sedation and mechanical ventilation during therapy (or at least during dressing changes), and the correct level of negative pressure and choice of continuous or intermittent pressure therapy. Moreover, this technique prescribes confinement in bed for days or weeks. Seven cases of cardiac rupture have been described during topical negative-pressure therapy in adults [21, 22]. Muscle flap is associated with delayed or primary closure. However, there are doubts about the functional and cosmetic results of major muscle flaps, with risk of breast hypoplasia and sensory loss occurring in adolescent girls [5]. The goal of muscle flap is to fill the sternal defect with well-vascularized healthy tissue. We believe that strong negative pressure after meticulous debridement achieves the same effect with a less aggressive approach. Finally, primary closure with short-term drainage advocated by the Ann Arbor group [8] is not very different
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
427
Table 7. Duration of Intravenous Antibiotic Therapy in Recent Studies of Mediastinitis in Pediatric Patients
References Sung et al [5] Ohye et al [8] Tortoriello et al [14] Al-sehly et al [16] a
Number of Patients
Duration of IV Antibiotic Therapy
Hospital Stay (days) Median and Range
11 57 15 29
4–6 weeksa 6 weeks 4–6 weeks 4–8 weeks
24 (8–140) 12 (4–88) 42.5 (16–163) N/A
6 weeks in patients with positive blood culture.
IV ⫽ intravenous;
N/A ⫽ not available.
from our policy. In their series, 3 of 42 patients (7%) were reoperated on for continuing sepsis. In our experience, the daily culture of mediastinal effluent has been efficient in preventing reoperation. This advantage is illustrated in the 3 patients with delayed sterilization followed by the rapid negative status of fluid culture after therapeutic modification. Our technique is associated with a short course of intravenous antibiotics, about 2 weeks versus 4 to 8 weeks most often recommended in the literature (Table 7). We have learned from our technique that mediastinal fluid remains infected a few days after surgical revision. The good results obtained with short-term chemotherapy are probably related to the efficiency of prolonging the drainage until the mediastinal effluent cultures negative for microorganisms. The outcome is a decreased hospital stay (a major factor in decreasing restraint owing to mediastinitis). In contrast to adult surgery, the prognosis of mediastinitis in pediatric surgery is much better with an overall success rate of approximately 100% [4, 5, 8, 14]. In most cases, death is not directly related to uncontrolled mediastinal sepsis [8, 16]. In this series, mortality was related to a patient’s associated disease and was not related to mediastinitis or therapeutic complications. Pneumococcus is uncommon in postoperative mediastinitis but associated with poor prognosis [23, 24]. In our patient, an extreme fragility of the inflammatory tissues was considered as the main cause of perioperative hemorrhage and death. The two other patients had well-documented nonseptic cause of death. Through a large series of patients we have demonstrated that primary closure with Redon catheter drainage is always efficient in the eradication of infection. Therefore, mediastinitis is not a direct cause of death in our experience. The modification of our initial therapy toward shorter drainage time and a shorter course of antibiotic therapy did not impair the good results obtained with this therapy.
References 1. Shumacker HB, Mandelbaum I. Continuous antibiotic irrigation in the treatment of infection. Arch Surg 1963;86:384 –7.
GENERAL THORACIC
Ann Thorac Surg 2007;84:423– 8
GENERAL THORACIC
428
ANSLOT ET AL POSTOPERATIVE MEDIASTINITIS IN CHILDREN
2. Ohuchi S, Kawazoe K, Ishihara K, Izumoto H, Eishi K. Management with closed irrigation for post-sternotomy mediastinitis: experience with the use of electrolyzed strong acid aqueous solution. Jpn J Thorac Cardiovasc Surg 2003; 51:511– 4. 3. Stiegel RM, Beasley ME, Sink JD, et al. Management of postoperative mediastinitis in infants and children by muscle flap rotation. Ann Thorac Surg 1988;46:45– 6. 4. Erez E, Katz M, Sharoni E, et al. Pectoralis major muscle flap for deep sternal wound infection in neonates. Ann Thorac Surg 2000;69:572–7. 5. Sung K, Jun TG, Park PW, Park KH, Lee YT, Yang JH. Management of deep sternal infection in infants and children with advanced pectoralis major muscle flaps. Ann Thorac Surg 2004;77:1371–5. 6. Agarwal JP, Ogilvie M, Wu LC, et al. Vacuum-assisted closure for sternal wounds: a first-line therapeutic management approach. Plast Reconstr Surg 2005;116:1035– 43. 7. Ramnarine IR, McLean A, Pollock JCS. Vacuum-assisted closure in the pediatric patient with post-cardiotomy mediastinitis. Eur J Cardiothorac Surg 2002;22:1029 –31. 8. Ohye RG, Maniker RB, Graves HL, Devaney EJ, Bove EL. Primary closure for postoperative mediastinitis in children. J Thorac Cardiovasc Surg 2004;128:480 – 6. 9. Durandy Y, Batisse A, Bourel P, Dibie A, Lemoine G, Lecompte Y. Mediastinal infection after cardiac operation: a simple closed technique. J Thorac Cardiovasc Surg 1989;97:282–5. 10. Calvat S, Trouillet JL, Nataf P, Vuagnat A, Chastre J, Gibert C. Closed drainage using Redon catheters for local treatment of poststernotomy mediastinitis. Ann Thorac Surg 1996;61: 195–201. 11. Kirsh M, Mekontso-Dessap A, Houel R, Giroud E, Hillion ML, Loisance DY. Closed drainage using Redon catheters for poststernotomy mediastinitis: results and risk factors for adverse outcome. Ann Thorac Surg 2001;71:1580 – 6. 12. Berg HF, Brands WG, Van Geldorp TR, KluytmansVandenBergh FQ, Kluytmans JA. Comparison between closed drainage techniques for the treatment of postoperative mediastinitis. Ann Thorac Surg 2000;70:924 –9. 13. Hulin S, Durandy Y. Postoperative mediastinitis due to methicillin-resistant Staphylococcus epidermidis with low sensitivity to vancomycin. Arch Pediatr 1997;41:1204 – 6.
Ann Thorac Surg 2007;84:423– 8
14. Tortoriello TA, Friedman JD, McKenzie ED, et al. Mediastinitis after pediatric cardiac surgery: a 15-year experience at a single institution. Ann Thorac Surg 2003;76:1655– 60. 15. Mehta PA, Cunningham CK, Colella CB, Alferis G, Weiner LB. Risk factors for sternal wound and other infections in pediatric cardiac surgery patients. Pediatr Infect Dis J 2000; 19:1000 – 4. 16. Al-sehly AA, Robinson JL, Lee BE, et al. Pediatric poststernotomy mediastinitis. Ann Thorac Surg 2005;80:2314 –20. 17. Long CB, Shah SS, Lautenbach E, et al. Postoperative mediastinitis in children: epidemiology, microbiology and risk factors for Gram-negative pathogens. Pediatr Infect Dis J 2005;24:315–9. 18. San Juan R, Aguado JM, Lopez MJ, et al. Accuracy of blood culture for early diagnosis of mediastinitis in febrile patients after cardiac surgery. Eur. J Clin Microbiol Infect Dis 2005; 24:182–9. 19. Fowler VG, Kaye KS, Simel DL, et al. Staphylococcus aureus bacteremia after median sternotomy: clinical utility of blood culture results in the identification of postoperative mediastinitis. Circulation 2003;108:73– 8. 20. Loos B, Kopp J, Kneser U, Weyand M, Horch RE. The importance of vacuum therapy in the treatment of sternal osteomyelitis from the plastic surgeons point of view. Zentralbl Chir 2006;131:124 – 8. 21. Sartipy U, Lockowandt U, Gäbel J, Jidéus L, Dellgren G. Cardiac rupture during vacuum-assisted closure therapy. Ann Thorac Surg 2006;82:1110 –1. 22. Abu-Omar Y, Naik MJ, Catarino PA, Ratnatunga C. Right ventricular rupture during use of high-pressure suction drainage in the management of poststernotomy mediastinitis [letter]. Ann Thorac Surg 2003;76:974. 23. Souza Neto EP, Celard M, Durand PG, Ninet J, Lehot JJ. Fulminant mediastinitis from Streptococcus pneumoniae following cardiac surgery. Ann Fr Anesth Reanim 2002;21: 603–5. 24. Maroto LC, Aguado JM, Carrascal Y, Rodriguez JE, Cortina JM, Rufilanchas JJ. Fulminant mediastinitis due to Streptococcus pneumoniae following cardiac surgery. Clin Inf Dis 1996;22:594 –5.