Interact CardioVasc Thorac Surg 2009;8:538-542. doi:10.1510/icvts.2008.187021
© 2009 European Association of Cardio-Thoracic Surgery
Institutional report - Cardiopulmonary bypass |
Circulating particles during cardiac surgery
Henrik Jönsson*,
Atli Eyjolfsson,
Sara Scicluna,
Per Paulsson and
Per Johnsson
Department of Cardiothoracic Surgery, Center for Heart and Lung Disease, Lund University Hospital, SE-221 85 Lund, Sweden
Received 25 June 2008;
received in revised form 14 January 2009;
accepted 19 January 2009
 Funding Sources: This study was funded by the Crafoord foundation in Lund, Sweden, and by Swedish governmental research grants (ALF funds).
*Corresponding author. Fax: +46 (46) 15 86 35.
E-mail address: henrik.jonsson{at}med.lu.se (H. Jönsson).
 |
Abstract
|
|---|
Shed blood is known to be a source of lipid micro-emboli in cardiac surgery. The aim of this study was to characterize the occurrence of these particles at different stages of the operation, and to study their occurrence in the circulation at multiple time-points after the retransfusion of shed blood. Forty-four patients undergoing routine surgery with cardiopulmonary bypass were included. Blood was sampled from the surgical field at different sampling locations during the operation. Shed blood was collected in a transfusion bag and retransfused. After which, blood was sampled from the arterial line of the heart-lung machine. A Coulter counter was used for particle determinion. The mean volume of shed blood collected was 340±215 ml. Particles in the size range 10–60 µm were found at varying concentrations, with the highest concentrations being found in blood collected after cannulation and from the pleura. After retransfusion of this blood, a biphasic response was seen in the blood drawn from the efferent line of the heart-lung machine. Particles are found in shed blood at all times during cardiac surgery, and when this blood was retransfused an increase was seen in particle concentration in the heart-lung machine.
Key Words: Particles; Lipid particles; Circulation; Shed mediastinal blood
|
1. Introduction
|
|---|
Cardiotomy suction blood collected from the surgical field during cardiac surgery with cardiopulmonary bypass (CPB) has been shown to be contaminated by lipid material [1, 2], that can form emboli in several organs [1–4]. In animal studies, a clear relation was found between lipid-laden retransfused blood and cerebral emboli [2], as well as other organs [5]. The findings in these experimental studies are consistent with findings of lipid deposits in the brain of patients who have undergone cardiac surgery [6–9].
Despite the debate on the relevance of these potentially harmful emboli, only a few studies have addressed the characteristics of this embolic phenomenon [1, 10]. This paper presents a descriptive analysis of this embolic phenomenon. The occurrence of embolic material was investigated at several sampling locations and events during cardiac surgery. In addition, after retransfusion of cardiotomy suction blood, the embolic response in the efferent part of the CPB-circuit was studied at multiple sampling times.
|
2. Material and methods
|
|---|
2.1. Subjects
After approval by the local ethics committee and informed consent, 44 patients undergoing only routine elective coronary artery bypass graft surgery (CABG) with CPB were included in the study.
2.2. Study design
During surgery, all cardiotomy suction blood was collected in a transfusion bag. After completing the distal anastomoses, the cross-clamp was replaced by a tangential occluder for the proximal anastomoses. At this time, the collected blood was measured and retransfused into the cardiotomy reservoir of the CPB circuit as quickly as possible, using a quick-prime set.
2.3. Blood and fluid sampling
A background sample was drawn from the peripheral arterial line after heparinization, but before cannulation. A sample of pericardial fluid was drawn immediately after the pericardium was opened. Samples of blood from the pericardium were drawn after cannulation, after removal of the aortic cross-clamp, and after decannulation but prior to protamine administration. In addition, a sample of blood was collected from the left pleural space at the end of the operation.
Before retransfusion of the collected shed blood, baseline samples were drawn from the CPB circuit and the transfusion bag. After retransfusion, samples were drawn from the CPB circuit proximal to the arterial in-line filter at 30, 60, 120, 150, 180, 300 and 600 s. All samples had a volume of 3 ml, and were analyzed with a Coulter counter immediately after the last sample was drawn.
2.4. Cardiopulmonary perfusion and surgical technique
Perfusion was performed with either a Stöckert S3 Heart-Lung machine with a Dideco Compactflo Evo Oxygenator and a 43-µm Cobe CV Sentry Arterial Filter (All Sorin group SpA, Milano, Italy), or a Jostra-Maquet HL20 Heart-Lung machine with a Quadrox oxygenator and a 40-µm Quart Arterial Filter (Maquet Critical Care, Solna, Sweden). Cardiotomy suctions were connected to a transfusion bag, and the collected blood was retransfused to the cardiotomy reservoir containing a 30-µm filter (Dideco Compactflo Evo) or a 40-µm filter (Jostra Maquet Quadrox). Surgery was performed in a standardized manner, where the left internal mammary artery (LIMA) was routinely harvested first and patients are then heparinized before pericardiectomy and LIMA is divided.
2.5. Coulter counter analysis
All blood samples and pericardial fluid were centrifuged at 4200 rpm for 45 min to separate blood cells and lighter particles (e.g. lipid particles). The supernatant was collected and stirred, and 400 µl of this plasma was diluted with 100 ml saline for analysis in the Coulter counter.
A MultisizerTM 3, Beckman Coulter Counter® with a 100-µm aperture probe (Beckman Coulter Inc, Fullerton, CA, USA) was used for particle size determination [11]. In the protocol used, 2 ml of the diluted specimen was analyzed at room temperature with a setting counting particles between 2 and 60 µm, at 0.2 µm intervals.
2.6. Data analysis
All data are presented as the mean±1 standard deviation (S.D.) or as a ±95% confidence interval (CI) was used. A Student's t-test with homoscedastic variance was used for group comparison.
The response in particles from samples taken from the efferent line of the CPB circuit was determined using area under the curve (AUC) during the entire sampling period. The background level (mean of the samples at 0 and 600 s) was subtracted from the values calculated in the interval 30–300 s.
|
3. Results
|
|---|
The highest number of particles larger than 10 µm was found in blood collected from the surgical field and from the transfusion bags (Fig. 1), and was significantly higher than the number of particles found in arterial blood before cannulation and in pericardial fluid after pericardiectomy (P<0.0001). The blood in the transfusion bags showed an abundance of particles in the range 10–60 µm, compared to that found in arterial blood before cannulation (Fig. 2). In addition, there was a significant difference between levels of particles in blood collected from the surgical field at different location points and that in the transfusion bag (P<0.0001, one-way ANOVA). A large interindividual variation was found in particle concentration in the transfusion bag blood (Fig. 3).
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 1. Mean (±95% CI) number of particles  10 µm at different sampling locations (arterial blood, pericardial fluid, pleural space and transfusion bag) and from the pericardium at different times during surgery (cannulation, removal of the cross-clamp and decannulation).
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 2. Size distribution (mean±1 S.D.) of particles found in the supernatant (after centrifugation) of arterial blood (•) and blood from transfusion bags ( ) grouped in 1 µm intervals.
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 3. Histogram showing the distribution of particle concentration found in blood from the transfusion bags for different patients.
|
|
After retransfusion of the blood from the transfusion bag, the concentration of particles in the arterial line in the CPB circuit showed a biphasic response (Fig. 4). The levels of particles 10 µm at 30, 60 and 90 s after retransfusion were significantly higher than before retransfusion (3562±5063, 3711±5416 and 2654±3004 vs. 1604±1593 particles/ml plasma, respectively, all P<0.001). However, no significant difference was found 120, 150, 180, 300 and 600 s after transfusion when compared with levels before retransfusion (2137±2650, 2369±2650, 3517±7701, 1556±2314 and 1705±2688 particles/ml, respectively). The levels of particles <10 µm were significantly increased at all times after retransfusion compared to that before transfusion (P<0.05 at 30, 60 and 600 s vs. 0 s, P<0.005 at 90, 120, 150, 180 and 300 s vs. 0 s).
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 4. Number of particles (10 µm) in the perfusion circuit at different points in time after retransfusion of cardiotomy suction blood (mean±95% CI). The number of particles is significantly different from that at baseline (0 s sample); *P<0.05 and **P<0.005.
|
|
Univariate analysis of the difference between the number of particles at the different location points during surgery and the response measured as the AUC only showed a significant correlation between the number of particles after cannulation and the AUC (r=0.41, P<0.05).
When testing perioperative variables (Table 1) and the number of particles in the transfusion bag using univariate analysis, only gender was found to be correlated with the number of particles. Higher numbers of particles (10 µm) were found in the transfusion bag blood collected from men than from women (122,287±98,537 vs. 27,250± 24,912 particles/ml plasma, P<0.005). No significant gender effect was seen on particle numbers at any of the other location points. No correlation was found between the concentration particles in the bag and the volume of the blood in the bag.
View this table:
[in this window]
[in a new window]
|
Table 1 Pre- and intra-operative variables. Hypercholesterolemia defined as treated with statins before surgery. Hypertension defined as diagnosed and treated with antihypertensive drugs
|
|
|
4. Discussion
|
|---|
The results of this study show that blood aspirated from the surgical field during cardiac surgery contains substantially higher amounts of particles larger than 10 µm than circulating arterial line blood. When this shed blood was retransfused to the CPB circuit an increase in the number of particles was observed in the arterial line of the CPB circuit.
The Coulter counter principle, albeit a well-proven technique for particle determination, has been used very little for the detection and characterization of emboli in cardiac surgery. Our group has previously reported the occurrence of these particles using a Coulter counter [10]. In that study, the supernatant containing particles was studied with gas chromatography, and a pattern of lipid distribution similar to that in mediastinal adipose tissue was found. Moreover, since the blood was centrifuged and the supernatant used for analysis in the Coulter counter, only embolic material with the same or a lower density than plasma will be analyzed. In addition, the results of this study, in terms of lipid particle size distribution are also consistent with the results of a study by Kaza et al., in which a lipophilic dye was used to detect lipid emboli [1]. It is therefore likely that the particles detected by the Coulter counter represent, at least to a majority, lipid particulate matter.
A large variation was found in the number of particles in the blood aspirated from the surgical field at different sampling locations (Fig. 1). Surprisingly, there was a considerable interpatient variation in particle concentration in the transfusion bags, as can be seen in Fig. 3. The same large variation was also seen in blood collected from other sampling locations. The highest levels were found immediately after cannulation and in the pleura, which could explain that this blood was the most in contact with fresh dissection surfaces in adipose tissue. The finding of very low levels of particles in the pericardial fluid immediately after pericardiectomy tells us that the particles do not originate from pericardial fluid, but are formed in the pericardium after surgery had begun.
We did not attempt to elucidate which patient factors determine the formation of particles in this study. The only perioperative variable that could be correlated with particle number was gender, and only at one sampling location. The reason for not finding other correlations is at least twofold. The first and most important, is that several of the potential factors that could influence emboli formation, such as total free lipid content in the surgical field, concentration of possible emulsifiers, temperature and energy added to the mixture that will facilitate emulsification, were not studied. The second explanation can be found in the large variation. Therefore, a larger number of patients would be needed to determine if there are any correlations. In addition, the sampling technique may also influence the representativeness of the samples. We choose to draw blood from the sampling-line in the CPB-circuit, because it is the easiest and safest sampling-point. Unfortunately, it is located before the arterial filter, which could influence results for larger particles. But the observation that there is a recirculation of larger particles indicates that filters may not act to remove these particles. If we look at the size profile of particles at the different sampling points after retransfusion, there is no shift towards smaller particles with time, implying that the larger particles also are recirculated. At present, the high variation must be considered as an observation, where only part of the variation can be explained in this study.
The biphasic response seen in the number of particles in the arterial line after retransfusion is interesting. A similar biphasic pattern was found in a porcine model in which lipid embolization was studied using radioactively labeled lipids [12]. In that study the first peak occurred after 40 s, and the second was observed after 8 min, and are consistent with the findings of this study. It was assumed that the second peak was a consequence of particle recirculation. Particle concentration response varied considerably on an individual level. The explanation of this variation is not obvious. Although it could be argued that biological systems vary, the sampling from the CPB circuit was performed in a standardized manner and the flow in the heart-lung machine varied little, thus making biological variation a less probable explanation. Another, more likely explanation, is that important points in time were missed in the sampling scheme used. With a flow in the CPB circuit of 4 l/min, 2 l of blood will pass through the system during 30 s with just 300 ml of blood being added rapidly, we could have missed the first peak in some individuals.
The technique used provides detailed information on both number and size distribution of particles. We omitted all particles smaller than 10 µm from the analysis for two reasons. First, particles smaller than 10 µm do not have the same potential to obstruct arterioles as larger particles. Secondly, in the size range between 2 and 10 µm other particles, such as residual erythrocytes, thrombocyte aggregates, chylomicrons and propofol emulsion, will interfere with the results. For instance, propofol emulsion has a mean diameter of 0.3 µm with particles up to 3 µm [13]. However, an abundance of particles smaller than 10 µm was found using the Coulter counter. Fig. 5 shows the number of particles in the 2–10 µm range and it can be seen that smaller particles are not cleared from the circulation as quickly as larger particles.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 5. Number of particles (<10 µm) in the perfusion circuit after retransfusion of cardiotomy suction blood (mean±95% CI). The number of particles is significantly different from that at baseline (0 s sample); *P<0.05 and **P<0.005.
|
|
That particles larger than 10 µm can be potentially harmful is obvious. Two studies have been conducted showing that emboli in this size range can actually cause cellular damage. In a recent study by Rapp et al., 50 µm particles were used in a rat model. They found MRI-changes, increased heat-shock protein 70 levels and histological findings revealed neuronal death in two of the six animals [14]. It should be noted that only one hundred 50 µm particles were infused. Omae et al. infused a larger number of particles (i.e. 1000 50 µm microspheres) in one carotid artery of a rat and found ipsilateral ischemic lesions at histological examination and impaired cerebral blood flow measured with MRI [15]. The numbers of particles used in these animal models should be compared to the concentrations found in the transfusion bag, which contained approximately 100,000 particles per ml plasma.
The results of the present study provide a suitable connection between the observation of lipid deposits in various organs after cardiac surgery, and the observation of lipid material in blood collected by means of cardiotomy suction during surgery. The study did not address the relationship between lipid particles and adverse neurocognitive outcome, or the specific origin of the particles measured. But, the observation of millions of particles entering the circulation will warrant further investigations on the topic.
|
References
|
|---|
- Kaza AK, Cope JT, Fiser SM, Long SM, Kern JA, Kron IL, Tribble CG. Elimination of fat microemboli during cardiopulmonary bypass. Ann Thorac Surg 2003;75:555–559.[Abstract/Free Full Text]
- Brooker RF, Brown WR, Moody DM, Hammon JW Jr, Reboussin DM, Deal DD, Ghazi-Birry HS, Stump DA. Cardiotomy suction: a major source of brain lipid emboli during cardiopulmonary bypass. Ann Thorac Surg 1998;65:1651–1655.[Abstract/Free Full Text]
- Appelblad M, Engstrom KG. Fat content in pericardial suction blood and the efficacy of spontaneous density separation and surface adsorption in a prototype system for fat reduction. J Thorac Cardiovasc Surg 2007;134:366–372.[Abstract/Free Full Text]
- Wright ES, Sarkozy E, Dobell AR, Murphy DR. Fat globulemia in extracorporeal circulation. Surgery 1963;53:500–504.[Medline]
- Bronden B, Dencker M, Allers M, Plaza I, Jonsson H. Differential distribution of lipid microemboli after cardiac surgery. Ann Thorac Surg 2006;81:643–648.[Abstract/Free Full Text]
- Moody DM, Bell MA, Challa VR, Johnston WE, Prough DS. Brain microemboli during cardiac surgery or aortography. Ann Neurol 1990;28:477–486.[CrossRef][Medline]
- Moody DM, Brown WR, Challa VR, Stump DA, Reboussin DM, Legault C. Brain microemboli associated with cardiopulmonary bypass: a histologic and magnetic resonance imaging study. Ann Thorac Surg 1995;59:1304–1307.[Abstract/Free Full Text]
- Challa VR, Moody DM, Troost BT. Brain embolic phenomena associated with cardiopulmonary bypass. J Neurol Sci 1993;117:224–231.[CrossRef][Medline]
- Hill JD, Aguilar MJ, Baranco A, de Lanerolle P, Gerbode F. Neuropathological manifestations of cardiac surgery. Ann Thorac Surg 1969;7:409–419.[Medline]
- Eyjolfsson A, Scicluna S, Johnsson P, Petersson F, Jonsson H. Characterization of lipid particles in shed mediastinal blood. Ann Thorac Surg 2008;85:978–981.[Abstract/Free Full Text]
- Kimelberg HK, O'Connor ER, Sankar P, Keese C. Methods for determination of cell volume in tissue culture. Can J Physiol Pharmacol 1992;70(Suppl):S323–S333.[Medline]
- Bronden B, Dencker M, Blomquist S, Plaza I, Allers M, Jonsson H. The kinetics of lipid micro-emboli during cardiac surgery studied in a porcine model. Scand Cardiovasc J 2008;42:411–416.[CrossRef][Medline]
- Park JW, Park ES, Chi SC, Kil HY, Lee KH. The effect of lidocaine on the globule size distribution of propofol emulsions. Anesth Analg 2003;97:769–771.[Abstract/Free Full Text]
- Rapp JH, Pan XM, Yu B, Swanson RA, Higashida RT, Simpson P, Saloner D. Cerebral ischemia and infarction from atheroemboli <100 microm in size. Stroke 2003;34:1976–1980.[Abstract/Free Full Text]
- Omae T, Mayzel-Oreg O, Li F, Sotak CH, Fisher M. Inapparent hemodynamic insufficiency exacerbates ischemic damage in a rat microembolic stroke model. Stroke 2000;31:2494–2499.[Abstract/Free Full Text]
|
|
Top
Abstract
1. Introduction
