Pediatr Cardiol (2007) 28:163–166

DOI 10.1007/s00246-006-0008-1

O R I G I N A L A RT I C L E

Mediastinitis in Pediatric Cardiac Surgery

Treatment and Cost-Effectiveness in a Low-Income Country

V. L. Vida Æ J. Leon-Wyss Æ A. Larrazabal Æ

S. Cruz Æ A. R. Castaňeda

Received: 8 September 2006 / Accepted: 18 November 2006

Springer Science+Business Media, LLC 2007

Abstract Mediastinitis is a serious postoperative Keywords Complications
Congenital heart disease

complication following pediatric cardiac surgery. The Infection
Surgery
objective of this study is to evaluate the cost-effective-
ness of surgical treatment for mediastinitis in Guate-
Mediastinitis is a serious postoperative complication
mala. All children who underwent a median
characterized by clinical and culture evidence of deep
sternotomy and developed postoperative mediastinitis
sternal infection involving the pericardial space. Its
between January 2004 and December 2005 were
reported worldwide incidence is between 0.1 and 5%
evaluated. Type of surgical treatment for mediastinitis,
[1, 2, 6, 10] of patients who undergo median sternot-
hospital outcome, and costs were analyzed. Eighteen
omy, and it presents a significant medical and financial
(3.3%) of the 535 children who underwent a median
burden [8–10]. Current surgical therapy involves a wide
sternotomy developed mediastinitis. Two patients
spectrum of treatment options, including delayed
underwent debridement of the infected tissues and
sternal closure, primary chest closure with continuous
delayed sternal closure, whereas 16 patients had
antibiotic irrigation or vacuum drainage, and muscle
debridement, primary chest closure, and continuous
flaps closure; however, there is no consensus about the
antibiotic irrigation of the mediastinum. All 11 patients
best treatment [2, 3, 10]. The objective of this study is
who had the diagnosis of mediastinitis within 2 weeks
to evaluate the cost-effectiveness of surgical treatment
after the operation survived. Three of the 7 patients
of children who developed postoperative mediastinitis
(43%) who had delayed diagnosis died (p = 0.0003); all
in Guatemala.
3 had osteomyelitis (p = 0.0007). Primary closure with
antibiotic irrigation was associated to a lower mortality
rate and proved less expensive in comparison to
delayed sternal closure (p = 0.003) mainly due to the Methods
shorter intensive care requirement. Debridement fol-
lowed by primary closure of the chest and continuous This study is a single-center, retrospective review of the
antibiotic irrigation of the mediastinum seems to be a medical records of 535 children who underwent
feasible and less expensive method to treat selected median sternotomy between January 2004 and Decem-
cases of postoperative mediastinitis in children. ber 2005. Demographic data of patients who developed
postoperative mediastinitis, type of surgical treatment
for mediastinitis, hospital outcome, and costs were
evaluated in the analysis.
V. L.Vida (&)
J. Leon-Wyss
A. Larrazabal
Actual hospital costs were calculated for every
S. Cruz
A. R. Castaňeda patient who underwent surgical treatment for postop-
Department of Pediatric Cardiac Surgery of Guatemala,
erative mediastinitis, considering only the additional
UNICAR, 9a Avenida, 8-00 Zone 11, Guatemala City,
CA, Guatemala hospital cost since the moment of the diagnosis of
e-mail: Vladimirovida@interfree.it mediastinal infection. Our cost-effective analysis

123
164 Pediatr Cardiol (2007) 28:163–166

compared only the direct costs (patient care related) in Table 1 Demographics and outcome of patients with postoper-
relation to the clinical success of the two procedures. ative mediastinitis
Hospital costs included operating room time, intensive Survivors Deaths p
care unit requirements, hospitalization, and antibiotic
therapy. To estimate the total cost for each patient, we No. of patients 15 3 —
Age, months 45.5 12.3 0.51
multiplied the unit cost of the various components of (SD; range) (46.1; 2–131) (9; 4–22)
care by the documented use and then summed the Time to diagnosis, 12.4 (4.1) 34 (17.5) 0.01
products. Medical, nursing, and technical staff salaries days (SD)
were considered in the total cost of the procedure for Osteomyelitis, n (%) 2 (13) 3 (100) 0.002
Nutritional statusa 91.3 (10.3) 77.3 (3) 0.04
every patient. Results are presented as mean and
a
standard deviation (median and range were used if Nutritional status: (current body weight/ideal body weight
according to height) · 100. Normal value: 90–110%.
data were not normally distributed). To assess if there
was a statistical difference between the groups, data
were analyzed using the two-sample Wilcoxon rank-
Table 2 Results of mediastinal cultures
sum (Mann–Whitney) test. Level of significance was
set at p £ 0.05. Type of pathogen No. of patientsa

Staphylococcus aureus 9
Klebsiella pneumoniae 3
Results Pseudomonas aeruginosa 2
Enterobacter sakazaki 2
Staphylococcus haemolyticus 1
Eighteen of 535 patients (3.3%) who underwent a
Staphylococcus saprophyticum 1
median sternotomy developed postoperative mediasti- Serratia rubidance 1
nitis. Median age of patients was 23 months (range, 2– Candida tropicalis 1
131 months). Mean time between the operation and Pseudomonas fluorescens 1
Pseudomonas filium 1
the diagnosis of mediastinitis was 16 ± 10.9 days
Candida albicans 1
(Table 1). Nine patients (50%) were readmitted after
a
previous uncomplicated hospital discharge. Initial Five patients had multiple–organism infection.
diagnostic signs included fever, leukocytosis, and ery-
thema of the incision in all 18 patients. Purulent
drainage from the incision occurred in 15 patients tinum after the debridement. Antibiotic irrigation was
(83%) and wound dehiscence in 11 patients (61%). discontinued after a median time of 8 days (range, 7–
The first 2 patients in our study underwent debride- 10), after the culture of the chest tube resulted negative.
ment of infected tissues and subsequent delayed sternal Systemic antibiotics were selected based on culture
closure with daily antibiotic irrigation of the medias- and sensitivity data and were administered for an
tinum and dressing changes while the chest was open. additional 6 weeks.
The remaining 16 patients underwent debridement of Twelve of 16 patients (75%) who underwent
infected tissues, immediate closure of the chest, and primary sternal closure with antibiotic irrigation were
continuous antibiotic irrigation of the mediastinum and extubated in the operating room and transferred
subcutaneous tissues. At the time of debridement, 5 directly to the ward. Six patients required prolonged
patients (27%) had an additional partial removal of the intubation, of which 4 cases were due to concomitant
sternal bone for osteomyelitis (Table 1). Mediastinal postoperative complications and 2 cases were due to
culture resulted positive in all 18 patients (Table 2). delayed sternal closure.
In patients who underwent primary closure, the Three of the 18 patients (16.6%) died of septic
antibiotic irrigation was done through two 3-mm silicon shock. Among the 2 patients who had delayed sternal
tubes inserted through a separate opening at the upper closure, 1 patient who had the diagnosis of mediasti-
part of the original incision: One irrigation tube was nitis 35 days after initial surgical repair died 8 days
placed in the mediastinum and one anterior to the after debridement. The other patient developed med-
sternum. Patients were initially irrigated with vanco- iastinitis 16 days after surgery, underwent successful
mycin (500 mg/1000 ml saline, 20 ml/hr), and two sternal closure 41 days after debridement, and had
patients received additional treatment with an antifun- multiple negative mediastinal cultures at the time of
gal agent for Candida species on culture (amphotericin chest closure. This patient was eventually discharged
B, 25 mg/1000 ml saline, 20 ml/hr). Three chest tubes home and did not present with recurrent mediastinal
were also placed within the pleural space and medias- infection.

123
Pediatr Cardiol (2007) 28:163–166
Two of the 16 patients (12.5%) who underwent
primary closure and antibiotic irrigation died. One
patient who had diagnosis of mediastinitis 51 days after
initial surgical repair died 7 days after debridement;
the other patient had recurrent mediastinal infection
14 days after initial debridement (Table 2) and died 2
days after a second debridement of infected mediasti-
nal tissue. The remaining 14 patients (87.5%) were
discharged home and they did not present with further
mediastinal infections.
All 11 patients who had the diagnosis of mediastinitis
within 2 weeks after the operation survived. Three
deaths occurred in the group of 7 patients (43%) who
had delayed onset of mediastinitis (more than 2 weeks, p
= 0.0003). All 3 patients had osteomyelitis (p = 0.0007)
and severe chronic malnutrition [4] (p = 0.04; Table 1).
Patients who underwent delayed sternal closure had
a significantly longer intensive care requirement (23.5 ±
23.3 vs 2.5 ± 4.9 days, p = 0.003) and hospital stay (31 ±
33.9 vs 14.1 ± 5.6 days, p = 0.04) and also had a higher
mortality rate (50 vs 12.5%) compared to patients who
underwent primary sternal closure (Table 1). Further-
more, the treatment cost for patients who had a delayed
sternal closure was significantly higher ($9,612 ± 9,298
vs $2,467 ± 1,486 for primary sternal closure; p =
0.003).The difference in costs between the two surgical
strategies was mainly attributed to the longer intensive
care unit requirement.
Discussion
Many strategies have been proposed to treat adult
patients who develop mediastinal infection following
cardiac surgery, including muscle flap closure, vacuum
systems, closed irrigation, and delayed sternal closure.
However, a consensus on the best treatment option for
the pediatric age group has not been reached [2, 3, 7,
10, 11]. Although the muscle flap closure treatment is a
valid option for adult patients with mediastinitis, there
are concerns about future musculoskeletal growth and
breast deformities in children [10]. Many of the other
treatment options are expensive and may not be
available in the setting of a Third World cardiac
surgery unit.
Our limited initial experience with debridement
followed by delayed sternal closure proved unsatisfac-
tory, mainly due to the excessive hospital costs, which
necessitated an alternative treatment. Therefore,
debridement followed by primary closure of the chest
and continuous irrigation of the mediastinum with
antibiotics was subsequently adopted. This proved to
be a reliable treatment option for postoperative med-
165
iastinitis, especially in children who were identified
early (<15 days from the primary median sternotomy),
with comparable survival rates [3, 8, 9]. In addition, the
majority of patients who underwent this technique
were extubated in the operating room and did not
require a prolonged postoperative intensive care unit
stay. Because the daily cost for intensive care unit stay
in our unit is six times more expensive than the daily
cost for a patient’s stay in the ward, this resulted in
significantly reduced hospital costs.
On the other hand, management of patients who
were diagnosed more than 15 days after surgery and
had sternal osteomyelitis at the time of debridement
remained less favorable. We believe that for this group
of patients in a developing country such as Guatemala,
better information for parents and primary care phy-
sicians about the possible risks of postoperative
complications, including mediastinitis, and early post-
operative followup will be helpful in the prompt
detection of mediastinal infection, which is essential
for successful surgical management.
In conclusion, our data suggest that early diagnosis
of mediastinal infection in children and primary
closure of the chest with continuous antibiotic irriga-
tion of the mediastinum may be associated with a
significantly lower medical and financial burden, which
is particularly important in countries with limited
health care resources. Evaluating the cost-effectiveness
of these procedures will facilitate budgetary realloca-
tion to efficiently treat more patients per year.
References
1. Al-Sehly AA, Robinson JL, Lee BE, et al. (2005) Pediatric
poststernotomy mediastinitis. Ann Thorac Surg 80:2314–2320
2. Barois A, Grosbuis S, Simon N, et al. (1978) Treatment of
mediastinitis in children after cardiac surgery. A study of 20
cases. Intensive Care Med 4:35–39
3. Berg HF, Brands WG, van Geldorp TR, Kluytmans-Vanden
Bergh FQ, Kluytmans JA (2000) Comparison between
closed drainage techniques for the treatment of postopera-
tive mediastinitis. Ann Thorac Surg 70:924–929
4. Cox J (2000) Nutrition. In: Siberry GK, Iannone R (eds) The
Harriett Lane Handbook. Mosby, Philadelphia, p 481
5. El Oakley RM, Paul E, Wong PS, et al. (1997) Mediastinitis
in patients undergoing cardiopulmonary bypass: risk analysis
and midterm results. J Cardiovasc Surg 38:595–600
6. El Oakley RM, Wright JE (1996) Postoperative mediastini-
tis: classification and management. Ann Thorac Surg
61:1030–1036.
7. Huddleston CB (2004) Mediastinal wound infections follow-
ing pediatric cardiac surgery. Semin Thorac Cardiovasc Surg
16:108–112.
8. Merrill WH, Akhter SA, Wolf RK, Schneeberger EW, Flege
JB (2004) Simplified treatment of postoperative mediastini-
tis. Ann Thorac Surg 78:608–612
123
166 Pediatr Cardiol (2007) 28:163–166

9. Ohye RG, Maniker RB, Graves HL, Devaney EJ, Bove EL 11. Zeitani J, Bertoldo F, Bassano C, et al. (2004) Superficial
(2004) Primary closure for postoperative mediastinitis in wound dehiscence after median sternotomy: surgical treat-
children. J Thorac Cardiovasc Surg 128:480–486 ment versus secondary wound healing. Ann Thorac Surg
10. Tortoriello TA, Friedman JD, McKenzie ED, et al. (2003) 77:672–675
Mediastinitis after pediatric cardiac surgery: a 15-year
experience at a single institution. Ann Thorac Surg 76:
1655–1660

123