| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2001;72:1327-1330
© 2001 The Society of Thoracic Surgeons
a Department of Anesthesiology, Gentofte Hospital, Copenhagen, Denmark
b Department of Clinical Microbiology, Herlev Hospital, University of Copenhagen, Copenhagen, Denmark
c Statens Serum Institut, Copenhagen, Denmark
Accepted for publication June 11, 2001.
Address reprint requests to Dr Schmidt, Department of Anesthesiology and Intensive Care, Odense University Hospital, DK-5000 Odense, Denmark
e-mail: henrik.schmidt{at}dadlnet.dk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentReferences |
|---|
Methods. Seventy-five patients undergoing electively performed CABG were included. Cultures of shed blood were taken at initiation of the autologous transfusion and the following morning. Infection variables were measured preoperatively and postoperatively. Infectious complications during the first postoperative week were registered.
Results. The frequency of patients with bacterial growth in the first culture was 0.22 (95% confidence interval: 0.12 to 0.31) compared with 0.04 (95% confidence interval: -0.044 to 0.087) in the second culture (p < 0.002). We found no significant difference in infection variables between patients with or without bacterial growth in the cultures. No patients suffered from early postoperative infectious complications.
Conclusions. There is no further contamination of the shed blood during the period between initiating the autologous transfusion and the following morning.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
A concern about the use of autologous blood transfusion after CABG is the sterility of the shed blood. For example, accidental bacterial contamination of the operation field during surgery might involve the shed mediastinal blood, implying a risk of bacterial growth in the shed blood and the reservoir containing the shed blood. Previous studies of autologous transfusion of shed mediastinal blood have found a varying degree of bacterial contamination of the blood [1, 4–7]. However, no study has examined the contamination of shed mediastinal blood any later than 6 hours postoperatively. Today, autologous transfusion is given for a length of time varying from patient to patient, depending on the amount of postoperative bleeding, and it does not seldom exceed 6 hours. Therefore, the aim of this study was to determine the degree of bacterial contamination of the shed mediastinal blood at initiation of the autologous transfusion after CABG and the following morning, before the removal of the mediastinal and pleural drains approximately 18 hours later. By this protocol, any further bacterial contamination and growth in the shed blood during the autologous transfusion period, no matter the duration of this, should be revealed. We also observed infection factors in the postoperative course to determine whether a bacterial contamination of the shed blood leads to an elevation in leukocytes, C-reactive protein (CRP), and procalcitonin (PCT).
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
With the exception of collecting the shed blood cultures and the preoperative blood samples, all patients were treated according to the department’s normal procedures for CABG, antibiotic prophylaxis, and preoperative and postoperative treatment.
Anesthesia was induced and maintained with fentanyl, midazolam hydrocloride, pancuronium bromide, and enflurane in oxygen. Hemodynamic variables were monitored with radial artery, central venous, and Swan-Ganz catheters. The operation was performed through a median sternotomy using standard techniques for cardiopulmonary bypass (CPB) with a crystalloid prime. CPB was conducted using an arterial roller pump with a Baxter cardiotomy/autotransfusion reservoir and a Baxter membrane oxygenator (Baxter Healthcare Corporation, Irvine, CA). St. Thomas’ cardioplegic solution (4°C) and ice-water (0°C) were used for myocardial protection, and CPB was carried out under moderate hypothermia (32°C to 36°C). Heparin-sodium (300 IU/kg) was used for anticoagulation and the activated time of coagulation was followed continuously throughout the operation. After CPB the heparin was neutralized with protamine sulfate. Surplus blood in the circuits was collected in bags for later retransfusion. Both pleural and mediastinal drains were connected to the Baxter cardiotomy/autotransfusion reservoir. The reservoir carried a 20-micron filter.
For antibiotic prophylaxis patients received cefuroxime (1.5 g three times a day) and gentamicin (3 mg/kg once a day for 2 days), with the first dose of both cefuroxime and gentamicin being given just after induction of anesthesia.
Postoperatively, patients were admitted to the intensive care unit where the autologous blood transfusion were initiated within half an hour after surgery. The shed mediastinal blood from the reservoir was transfused every hour until the first postoperative morning—approximately 18 hours later—if more than 20 mL of blood had accumulated. Autologous transfusion was discontinued after 3 hours without bleeding if the blood coagulated in the autotransfusion reservoir or if the patients developed massive hematuri. Thereafter the reservoir served only as a receptacle for the shed blood until the drains were removed from the patients the following morning. The autologous transfusion system was a closed circuit.
If patients were reoperated on before the first postoperative morning, the autologous transfusion was not restarted after the second surgical procedure and the patients were excluded from the study.
When initiating the autologous transfusion, the first set of blood cultures—including two aerobic and two anaerobic Bactec (Becton Dickinson, Sparks, MD) blood culture bottles with resin with 40 mL blood in total—was obtained from the reservoir. A second similar set of cultures was obtained on the morning of the first postoperative day before the removal of the pleural and mediastinal drains approximately 18 hours later.
The bottles were sent to the local department of clinical microbiology where culture, typing, and determination of antibiograms were done following the standard procedures of the department. The bottles were cultured for as long as 7 days. To reveal any accidental contamination of the shed blood cultures originating from the person obtaining the samples, skin cultures were taken from that person’s nostrils, fingers, and armpits.
A number of patient variables were recorded including sex, age, bypass time, millilitres of autologous blood transfusion, and the overall amount of postoperative bleeding. Infectious complications during the first week, eg, mediastinitis or sepsis, were recorded by reviewing the patients’ case records.
Blood samples for analysis of leukocytes, CRP, and PCT were taken from all patients immediately before the operation, on initiation of autologous transfusion, on the morning of the first postoperative day, on the second postoperative day, and on the fourth postoperative day. For determination of PCT concentrations, serum was stored at -20°C until analysis. The PCTsanalyses were done with the help and equipment of BRAHMS Diagnostica (Berlin, Germany).
Samples for analysis of the concentration of gentamicin and cefuroxime were taken from the circulating blood and the shed mediastinal blood from 10 patients. The samples were taken when initiating the autologous transfusion, on the morning of the first postoperative day, on the second postoperative day (only circulating blood), and on the fourth postoperative day (only circulating blood). The samples for analysis of the concentration of gentamicin were sent to the local department of clinical microbiology and analyzed according to standard procedures of the department. For determination of the concentration of cefuroxime, the samples were centrifuged, serum was stored at -20°C until analysis was performed at the State Serum Institute, Copenhagen, Denmark, using a biologic agar disk diffusion assay.
The protocol was sent to and accepted by the local ethic committee (KA 98116). All patients gave their informed consent before entering the study.
Statistics
We used nonparametric statistics including the Mann-Whitney U test, the Kruskall-Wallis test, and the 
2 test. A p value less than 0.05 was considered to indicate significance. The results are presented as the median value with the first and third quartile.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
Patients were post hoc divided into two groups: group A consisting of 55 patients without any bacterial growth in the shed blood cultures, and group B consisting of 18 patients with bacteria in one or more of the eight culture bottles obtained from each patient. In group B, 1 patient showed bacterial growth in both the first and the second set of cultures, whereas 15 other patients only showed growth in the first set of cultures and 2 patients solely in the second culture set. The frequency of patients with growth in one or more culture bottle were 0.22 (95% confidence interval: 0.12 to 0.31) in the first set of cultures compared with 0.04 (95% confidence interval: -0.0044 to 0.087) in the second set (p < 0.002).
We found 35 different isolates in the culture bottles with growth. The bacteria were Staphylococcus epidermidis, other coagulase negative staphylococci (CNS), and a single isolate of diphtheroids. Eight isolates of CNS were sensitive to gentamicin and to dicloxacillin and hence to other ß-lactam antibiotics as well, eg, cefuroxime. Seven isolates showed resistance to dicloxacillin but were sensitive to gentamicin, whereas 18 isolates were resistent to both dicloxacillin and gentamicin. Two isolates from 2 different patients did by accident not undergo determination of resistance. The antibiograms of the bacteria isolated from the shed blood were, with the exception of one isolate, not in agreement with that of the skin inhabitants of the person obtaining the cultures.
There were no significant difference between patients with and without bacterial growth regarding preoperative, perioperative, and postoperative characteristics (Table 1). Table 2 shows the concentration of antibiotics in serum and in the shed blood. Table 3 shows the concentrations of leukocytes, CRP, and PCT preoperatively and postoperatively.
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
We found a significantly larger number of patients with positive cultures at the beginning of the autologous transfusion compared with the first postoperative morning, approximately 18 hours later. Thus, bacteria were indeed found when culturing the shed blood but they vanished in the period between the first and second culture. Our concern about further bacterial contamination of and growth in the shed blood during the autologous transfusion was therefore not confirmed by our results. As far as bacterial growth in the shed blood is concerned, there is apparently no obstacle to autologous transfusion exceeding a 6-hour period, ongoing up to approximately 18 hours. However, because of the low incidence of early postoperative infections in general, we cannot make conclusions regarding a relationship between bacterial growth in shed blood cultures and subsequent clinical infection in the recipient. This would require a study with a very large number of patients.
Previous studies on retransfusion of shed mediastinal blood after heart surgery have shown a varying degree of bacterial contamination. Solem and associates [4] obtained aerobic and anaerobic cultures of shed blood from 20 patients at 2, 4, and 6 hours postoperatively. Three out of 120 turned out positive. Adan and colleagues [5] did not recover any bacteria at all in shed blood cultures from 25 patients 4 hours postoperatively. Thurer and colleagues [6] isolated bacteria from 3 out of 49 patients whereas Lepore and coworkers [7] found a frequency of positive cultures of 19% after 6 hours autologous transfusion and Schaff and coworkers found growth in 17% of the cultures [1].
The variation in bacterial growth between these studies and ours might be explained by different volumes of shed blood used for culturing, since the sensitivity towards detecting bacteria is increased by using larger amounts of blood for culturing [8]. Only Adan and colleagues [5] specify the blood volume being used for culturing in their study, that is, 10 mL in total compared with our 40 mL of shed blood per culture. Different blood culture systems with different sensitivity is another possible explanation to the variation in bacterial growth.
In several of the studies quoted above, the isolated CNS and diphtheroids are interpreted as being contaminants introduced through nonsterile aspiration when obtaining the cultures or from the laboratory. Because of the large significant difference between the number of patients with bacterial growth in the first and in the second culture, we do not find this likely. Furthermore, none of the antibiograms of the isolated bacteria except one were matching that of the skin bacteria of the person obtaining the cultures. Since most of the isolated bacteria show resistance to ß-lactam antibiotics or gentamicin or both, we believe that they originate from hospital settings with a large use of these antibiotics, the bacteria possibly being introduced to the open mediastinum during surgery either from a variety of sources in the operating theater [9–12] or from the patient’s skin. Most of the patients had spent several days at the hospital preoperatively and had time to acquire the hospital flora. Postoperatively, the autologous transfusion system formed a closed circuit, prohibiting any further contamination of the mediastinum and the shed blood from the surroundings. As bacteria could not be retrieved in the second set of shed blood cultures, the patient’s immune system and the antibiotics seem to be able to combat any bacteremia or bacteria present in the shed blood and the reservoir after initiating the retransfusion. The measurements of antibiotics in serum and in shed blood confirm that the patients receive antibiotics in amounts adequate for susceptible bacteria and that the antibiotics actually can be found in the shed blood. However, the antibiotics might interfere with the growth of the bacteria when culturing the shed blood. It is unclear to what degree the resin in the blood culture bottles inactivate the antibiotics. We therefore cannot totally exclude the possibility of the shed blood being contaminated with bacteria other than those isolated.
Bacterial growth did not have an effect on infection variables in the postoperative course. An inflammatory response to the surgical trauma and cardiopulmonary bypass is probably the explanation to the increase in both leukocytes, CRP, and PCT postoperatively. In summary, our data are consistent with the conclusion that there is no increase in the bacterial contamination of the shed mediastinal blood in the period between initiation of autologous transfusion and the following morning, indicating that there is no hindrance to autologous transfusion of shed mediastinal blood for more than 6 hours and as long as 18 hours, as far as bacterial contamination is concerned.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentReferences |
|---|
degran K. Autotransfusion of mediastinal blood in cardiac surgery. Scand J Thor Cardiovasc Surg 1989;23:47-49.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S. Dial, D. Nguyen, and D. Menzies Autotransfusion of Shed Mediastinal Blood: A Risk Factor for Mediastinitis After Cardiac Surgery? Results of a Cluster Investigation Chest, November 1, 2003; 124(5): 1847 - 1851. [Abstract] [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 |
