| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1999;117:496-505
© 1999 Mosby, Inc.
SURGERY FOR CONGENITAL HEART DISEASE |
From Schneider Children's Hospital of the Long Island Jewish Medical Center,a University of Michigan Medical Center,b and the Albert Einstein College of Medicine,c New Hyde Park, NY.
Read at the Seventy-eighth Annual Meeting of The American Association for Thoracic Surgery, Boston, Mass, May 3-6, 1998.
Received for publication May 8, 1998. Revisions requested June 11, 1998. Revisions received Nov 12, 1998. Accepted for publication Nov 12, 1998. Address for reprints: Hani A. Hennein, MD, Section of Pediatric Cardiothoracic Surgery, Schneider Children's Hospital of the Long Island Jewish Medical Center, Albert Einstein College of Medicine, New Hyde Park, NY 11040.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsDiscussionConclusionsAppendix: Discussion References |
|---|
, interleukin-1ß, interleukin-6, interleukin-8, and interleukin-10) data were collected for statistical analysis. |
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionConclusionsAppendix: Discussion References |
|---|
1 both of which may be removed by modified ultrafiltration (MUF). 2 Venovenous MUF is a further modification of this technique, which has potential advantages of shear simplicity and avoidance of an obligatory shunt. |
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionConclusionsAppendix: Discussion References |
|---|
Anesthetic techniques
Patients older than 1 year received oral midazolam (0.7 mg/kg) for premedication. Anesthesia was induced and maintained with a continuous infusion of 2 µg x min–1 x kg–1 midazolam, 2 µg x min–1 x kg–1 alfentanil, and 120 µg x min–1 x h–1 vecuronium. All patients were intubated orally with an uncuffed endotracheal tube. Standard monitoring included continuous arterial and central venous blood pressure measurement, pulse oximetry, and nasopharyngeal and bladder temperature. Perioperative antibiotics consisted of cefazolin 35 mg/kg administered every 8 hours for the first 48 hours after the operation. No patient received aprotinin or corticosteroids.
CPB
Patients were placed on CPB with bicaval cannulation and a single aortic cannula. CPB was conducted with a roller pump (model GA-9901; Stöckert Instrumente GmbH, Munich, Germany), and gas exchange was achieved with a hollow-fiber membrane oxygenator (Dideco Lilliput D901; Dideco Masterflow, Sorin Biomedical, Irvine, Calif). A 40-µm arterial line filter (Pall LPE 1440; Pall Biomedical, East Hills, NY) was used in all patients. The circuit was primed with Plasma-Lyte A solution (Baxter Healthcare, Deerfield, Ill), and sufficient packed red blood cells were added to achieve a hematocrit value of 20% to 25%. CPB was performed with deep-to-moderate hypothermia (temperature 20°C to 32°C), with the alpha-stat method for blood gas regulation. Flow rates of 25 to 150 mL/kg per minute were used according to the intraoperative conditions, and mixed venous oxygen saturation was continuously followed (CDI 100; 3M Health Care, Tustin, Calif).
Anticoagulation was achieved with an initial bolus of heparin sodium (250 IU/kg; Elkins-Sinn Inc, Cherry Hills, NJ) injected into the right atrium before cannulation, followed by a continuous infusion of 62.5 IU x kg–1 x h–1 until the end of CPB. The activated clotting time was maintained over 450 seconds throughout the CPB period. The heparin was reversed with 3.5 mg/kg protamine sulfate (Eli Lilly Co, Indianapolis, Ind) administered over 3 to 5 minutes to achieve an activated clotting time that approximated the baseline.
Myocardial preservation was achieved with cold blood cardioplegic solution delivered by a modified cardioplegia system (Vanguard BCD PTS 7004; Sorin Biomedical, Irvine, Calif). Cardioplegic solution was infused into the aortic root at a pressure not exceeding 120 mm Hg at a flow rate of 80 to 120 mL/min. The initial dose was 30 mL/kg, with subsequent doses of 15 mL/kg at 20-minute intervals.
Autotransfusion of shed blood during the operative procedure was used for patients weighing more than 30 kg and in all reoperations. Autotransfusion was performed with a cell separator system (Haemonetics Corporation, Braintree, Mass) with a small-volume bowl to wash and centrifuge aspirated blood. Autotransfusion of shed blood was used independently of the patient group to which the patient was randomized.
CUF was performed according to the circuit diagrammed in Fig. 1, A. The ultrafilter was positioned in the bypass circuit with its inlet distal to the oxygenator and its outlet directed toward the venous reservoir. CUF was carried out during the rewarming phase of the CPB period, after the patient's nasopharyngeal temperature reached approximately 30°C. The target volume for ultrafiltrate removal was the priming volume plus any additional fluid administered during the CPB period. Suction was applied to the filtrate port at a pressure not exceeding 200 mm Hg, and ultrafiltrate was removed at a rate not exceeding 50 mL/kg per minute. As has been previously reported, 3 CUF tends to be inconsistent in achieving this goal in clinical practice.
|
, B. The inferior vena caval cannula was used as the inflow to the ultrafilter, and blood was returned to the patient via the cardioplegia line attached to the superior vena caval cannula. To convert from CUF to venovenous MUF, the blood supply line to the cardioplegia device was disconnected from the arterial blood source and connected to a stopcock on the venous line near the venous reservoir (Fig. 1, A and B). The purge line from the cardioplegia device to the cardiotomy was closed and removed. After flushing was completed, the cardioplegia infusion line is connected to the inferior vena cava (IVC) cannula by a Luer-Lok connector. When single venous cannulation was used for CPB, a vessel cannula (DLP model 30001; Medtronic Inc, Grand Rapids, Mich) was attached to the cardioplegia line and inserted directly into the atrium for return to the patient. Returned blood was rewarmed after ultrafiltration with the heat exchanger of the cardioplegia device before being returned to the patient. A Hemocor HPH400 (Minntech Corporation, Minneapolis, Minn) was used for patients weighing less than 30 kg, and the Hemocor 1000 (Minntech Corporation) was used for patients weighing more than 30 kg. As with the use of CUF, suction was applied to the filtrate port at a pressure not exceeding 200 mm Hg; ultrafiltrate was removed at a rate not exceeding 50 mL/kg per minute, and target volume for ultrafiltrate removal was the priming volume plus any additional fluid administered during CPB.
Blood sampling
Blood samples were collected from the arterial catheter before CPB and after the operation and from the oxygenator during extracorporeal circulation. Samples were collected directly after induction of anesthesia (T1), 3 minutes after commencement of CPB (T2), 3 minutes after aortic crossclamping (T3), directly before the administration of protamine (T4), during skin closure (T5), and at 2 (T6) and 24 hours (T7) after the operation. Blood samples were collected into sterile vacuum tubes containing 0.5 mL ethylenediamine tetraacetic acid and immediately centrifuged. An aliquot of plasma was obtained and stored at –72°C until the cytokine assays were performed.
Cytokine assays
IL-1ß and IL-10 levels were measured by enzyme-linked immunosorbent assay with matched antibody pairs (R&D Systems, Minneapolis, Minn, and Pharmigen, San Diego, Calif, respectively). The minimum detectable levels was 30 pg/mL for IL-1ß and 0.975 pg/mL for IL-10. IL-8 concentrations were determined with a specific enzyme-linked immunosorbent assay developed and reported on by our laboratory. 4 The minimum detectable concentration for IL-8 was 30 pg/mL. IL-6 levels were likewise determined by a bioassay developed in our laboratory. 4 The minimum detectable level for IL-6 was 1.5 pg/mL. Tumor necrosis factor– (TNF-) was measured with the WEHI 164 subclone 13 cell line as previously reported. 4 The minimum detectable level for TNF- was 1.5 pg/mL.
Statistical analysis
Statistical analyses were run with commercially available software (SAS version 6.08; SAS Institute Inc, Cary, NC) on personal desktop computers (IBM Corporation, Armonk, NY). Mean values were reported as the mean ± SEM, and comparison of 2 mean values was performed with Student t test analysis. Multiple means were compared with analysis of variance (ANOVA) and grouped according to Duncan's multiple range test. 5 Categoric variables were analyzed with 
2 analysis with P values reported according to Fisher's exact 2-tailed test. 5 Variables that were significant at the P 
.05 level by univariate analysis were entered into a stepwise multiple regression analysis to test for statistical independence, assuming statistical independence at a P value .05.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionConclusionsAppendix: Discussion References |
|---|
|
). A small volume of ultrafiltrate was removed in the venovenous MUF group as a consequence of the preparation, de-airing, and priming of the venovenous MUF circuit. This volume (3.2 ± 2.5 mL/kg) was added to the ultrafiltrate volume reported for the venovenous MUF group.
|
). The hematocrit value in this group reflects the effectiveness of CUF in hemoconcentrating the priming volume during the CPB period. Conversely, the lowest hematocrit value was in the control group, which fell to approximately one third of the preoperative level (P = .0001). Patients in the venovenous MUF group were midway between CUF and control groups.
Patients in the venovenous MUF group had the highest postoperative hematocrit level, ending up with a value somewhat higher than their preoperative level (Table II). These results were significantly different from either the CUF or control groups.
Cytokine levels
TNF- levels rose in control and venovenous MUF groups but not in patients undergoing CUF (Fig 2). TNF- levels peaked at the time of skin closure and were significantly lower in the CUF group as compared with the control or venovenous MUF groups (40.2 ± 7.3 pg/mL vs 88.0 ± 15.2 pg/mL and 69.1 ± 10.5 pg/mL for control and MUF groups, respectively; P = .02). Mean TNF- levels (mean of all TNF- levels during the course of the study period) were also significantly lower in the CUF group as compared with either the venovenous MUF or control groups (40.5 ± 6.2 pg/mL vs 71.0 ± 6.9 pg/mL and 58.9 ± 5.21 pg/mL for control and MUF groups, respectively; P = .003).
|
levels, peak IL-1ß levels were highest in patients undergoing CUF (Fig. 3) and lowest in control patients. IL-1ß levels peaked at the time of skin closure, at which time the control values were significantly lower than the CUF group (21.2 ± 6.0 pg/mL vs 67.4 ± 9.8 pg/mL; P = .05). Mean IL-1ß levels, however, were not significantly different among the 3 groups.
|
IL-8 levels varied widely during the course of the study period, peaking at the time of skin closure. Peak and mean IL-8 levels were not significantly different among the 3 groups.
Anti-inflammatory cytokine IL-10 levels rose in all 3 groups of patients, peaking at the time of protamine administration (Fig. 4). Patients who underwent either CUF or venovenous MUF had rapid clearances of IL-10, although control patients maintained elevated levels that lasted into the early postoperative period. At both 2 and 24 hours after operation, IL-10 levels were significantly greater in control patients as compared with patients who had undergone either CUF or venovenous MUF. Mean IL-10 levels were significantly higher in the control patients as compared with patients who had undergone either the venovenous MUF or CUF (1.26 ± 0.35 pg/mL vs 0.42 ± 0.07 and 0.61 ± 0.09 pg/mL for venovenous MUF and CUF, respectively; P = .02).
|
Patients who underwent venovenous MUF had the shortest duration of postoperative intubation, followed by those who underwent CUF and then by those in the control group (Table III). The 3 groups were statistically distinct from each other, suggesting that venovenous MUF was more effective than CUF in shortening the duration of mechanical ventilation.
|
). Reduced chest tube drainage translated into reduced use of blood products in the postoperative period, with patients in the venovenous MUF group having the least amount of transfused blood products.
Patients who underwent venovenous MUF had the shortest duration of stay both in the intensive care unit (ICU) and in the hospital (Table III). For ICU days, the venovenous MUF and CUF groups were both statistically different from the control group. For hospital stay, the 3 group were statistically distinct, suggesting that venovenous MUF was superior to CUF and CUF was superior to no ultrafiltration in reducing the duration of the overall convalescence.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionConclusionsAppendix: Discussion References |
|---|
2,3; improving hemodynamics, 8,9 ventricular function, 10,11 and cerebral recovery 12; and reducing postoperative morbidity 13 after CPB. Venovenous MUF is a modification of the technique in which blood is removed from the systemic venous system, ultrafiltered, and returned to the same venous system. One of the most important advantages of venovenous MUF is that it does not cause an obligatory left-to-right shunt during the immediate period after CPB. Depending on the amount of shunting, arteriovenous MUF may steal a significant portion of the cardiac output and volume overload the ventricle, resulting in increased myocardial oxygen demand coupled with decreased coronary perfusion and delivery. Such hemodynamic considerations are of particular importance in neonates and patients with decreased myocardial reserve and in patients undergoing staged palliation for single ventricle morphologic examination.
For these reasons, venovenous MUF was developed as a means of removing extracellular volume without an obligatory left-to-right shunt by returning the reduced volume to the same physiologic cardiac chamber. Whereas venovenous MUF is not contraindicated in any group of patients, its use requires consideration for the potential of recirculation and must be thoughtfully applied in patients undergoing a superior cavopulmonary anastomosis.
Recirculation
Early on in our experience, we were concerned about whether recirculation from 1 venous cannula to the other would have an adverse effect on the efficiency of venovenous MUF, much as may occur during venovenous extracorporeal membrane oxygenation. 17 However, because 2 separate venous cannulas are used after CPB, as opposed to the single 2-stage cannula used in extracorporeal membrane oxygenation, recirculation has not been of any consequence in our patients.
Superior cavopulmonary anastomosis
After a superior cavopulmonary anastomosis, such as in a bidirectional Glenn or hemi-Fontan procedure, the direction of the flow in the venovenous MUF circuit must be considered. The IVC cannula must be the inflow to the hemoconcentrator, although the superior vena cava cannula must be the return cannula to the patient. A reversal of this setup (ie, superior vena cava to hemoconcentrator to IVC) will steal blood away from the pulmonary circulation, resulting in severe cyanosis.
Venovenous versus CUF
MUF was consistently a more effective means of removing volume as compared with CUF (Table II). There were 2 reasons for this: CUF was limited to the rewarming phase of CPB, and CUF was terminated whenever the blood volume in the CPB circuit became too low. Conversely, venovenous MUF was conducted after CPB under no such constraints, and because much of the CPB circuit is bypassed with this technique, a much lower circulating volume is required for ultrafiltration (Figs. 1, A and B).
Clinical consideration
The greater efficiency of MUF over CUF has been reported in both prospective randomized studies 2 and in retrospective studies. 13 These and other studies 18 have demonstrated a shorter duration of ventilator dependence, postulated to be due to a reduced impairment in oxygenation seen after a systemic inflammatory response. 19 Although not all studies are consistent in agreement, 10 the balance of studies suggests that MUF is superior to CUF in improving postoperative convalescence.
The reduction in total body water seen in our patients (Table III) was similar to that reported in patients undergoing arteriovenous MUF 2 and may account for the reduction in ventilator dependence and the length of convalescence in this group of patients. In addition, chest tube drainage was less in patients undergoing venovenous MUF, translating into reduced usage of blood products (Table III), corroborating several studies that report similar findings. 2,7,8,13
Inflammation after CPB
The release of inflammatory cytokines has been consistently demonstrated after CPB. 14,16,18 Inflammation, along with an increase in total body water, may be particularly deleterious in neonates and infants in whom capillary membranes are variably developed. Ultrafiltration has been reported by several authors 14-16,18 to remove some of the inflammatory mediators, although these findings are not entirely consistent.
TNF- has been shown to increase after CPB in many studies 21-24 and has been associated with decreased myocardial function and multisystem organ failure. In the present study, TNF- rose during CPB in the venovenous MUF and control groups, consistent with removal of TNF- by both venovenous MUF and CUF. IL-1ß also increased during CPB but was highly variable as compared with TNF-. Other studies 22,25 have demonstrated an increase in IL-1ß after CPB, implicating this cytokine in the pathogenesis of multisystem organ damage. Increased levels of IL-6 and IL-8 have been more consistently associated with CPB, and although their levels generally correlate with the duration of CPB, no hemodynamic effects have been ascribed to them. 21-23,26 The present study corroborates the increase in both IL-6 and IL-8 after CPB.
The release of the anti-inflammatory cytokine IL-10 is unique in that it may play a protective role by suppressing the production of the proinflammatory cytokines. 18,27,28 IL-10 levels rose into the early postoperative period in patients who did not undergo ultrafiltration (Fig. 4). Patients who underwent ultrafiltration had comparably lower IL-10 levels than those who did not, and these levels reached baseline within 2 hours of the CPB period. These patterns are consistent with clearance of IL-10 by both venovenous MUF and CUF.
In general, cytokines generated in response to CPB were removed to varying degrees by ultrafiltration. Whereas removal of proinflammatory cytokines may be seen as desirable, removal of anti-inflammatory cytokines may not. The net balance appears to be favorable, as evidenced by the improved clinical status of the subject patients.
Limitations of the study
The technique of ultrafiltration makes a total blinding to the operating room team impossible. In particular, the operative procedure and extracorporeal circulation was performed to maximize benefit and minimize risk to the patient.
The study may be limited by the relatively low number of patients in each group and by the variety of operative procedures performed. Nonetheless, the statistical analysis of both clinical variables and individual cytokines reached statistical significance, underscoring the clinical importance of these interventions.
|
Conclusions |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionConclusionsAppendix: Discussion References |
|---|
|
Appendix: Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionConclusionsAppendix: Discussion References |
|---|
With the circuit used at Great Ormond Street and at The Children's Hospital of Philadelphia, it is also possible to concentrate and reinfuse the circuit volume through the venous cannula without interrupting the MUF process. MUF has been shown to remove excess water and small molecular weight solutes, including many inflammatory mediators; and its use has been shown to reverse hemodilution, decrease the need for transfusion, increase lung compliance, and, importantly, improve hemodynamics and left ventricular function after CPB.
Similar studies at many institutions have demonstrated that alterations in the way ultrafiltration is performed both during and after bypass may increase the benefits of this technique. Studies such as the one just presented provide important evidence concerning the optimal use of these techniques. The authors have demonstrated that the use of venovenous MUF is safe and effective after CPB in children. They demonstrate a significant filtration of water and inflammatory mediators compared with control patients and those patients undergoing CUF.
Benefits of venovenous ultrafiltration included increased hematocrit value, decreased bleeding, decreased duration of mechanical ventilation, and decreased hospital stay; however, they do not report their hemodynamic data and, importantly, venovenous MUF was not compared with standard MUF in this trial.
The authors evaluated venovenous MUF because of concerns over possible adverse effects of standard arteriovenous ultrafiltration. The authors state that this may be a potential problem and result in hemodynamic instability, particularly in neonates. Studies in London and at The Children's Hospital of Philadelphia, however, have documented the safety and efficacy of standard MUF, even in critically ill neonates. Almost no adverse events have been described or reported during the performance of MUF, and indeed, a consistent finding is a significant and sustained rise in blood pressure and cardiac output during the period of ultrafiltration. We presented data at the meeting of The Society of Thoracic Surgeons in 1997 documenting the safety and efficacy of standard arteriovenous MUF in infants after stage I palliation for the hypoplastic left heart syndrome.
Do you have documentation of adverse effects of the left or right shunt during arteriovenous MUF or is this merely a theoretic concern?
You hypothesize that the technique of venovenous MUF will be particularly valuable in neonates; however, in this study the mean age of the patients is approximately 8 years and the mean weight is greater than 20 kg. The benefits of standard MUF have been shown to be the greatest in the smallest infants. How many neonates have undergone venovenous MUF? Are the benefits the same?
Did you assess hemodynamic status during the period of venovenous MUF, and does this technique provide the same hemodynamic benefits as arteriovenous MUF? Have you compared your results with those of standard arteriovenous MUF? Before this technique can be recommended, it must be shown to be as effective as standard MUF. There must be some obligatory degree of recirculation between the venous cannula, and this may decrease the efficiency.
Finally, as you know, Journois' colleagues from Paris have reported the use of zero balance ultrafiltration in which ultrafiltration is performed both during and after separation from bypass, and they have shown that this is superior to either technique alone. The data in this study suggest that there is an additional benefit to ultrafiltration both during rewarming and after CPB. What is your current technique of ultrafiltration and do you combine the 2 techniques?
Dr Hennein. I would first like to acknowledge Dr Gaynor's important work in this field in showing that there is improved ventricular function with arteriovenous MUF. He does raise some very important questions regarding the comparison of arteriovenous with venovenous MUF.
We have not performed arteriovenous ultrafiltration because of our concerns about its use in the neonatal patient. For this group of patients, we follow more theoretic principles rather than actually proving that arteriovenous MUF has any untoward effects.
Our logic goes like this: After CPB, during a reasonably tenuous time in the operative procedure, to actually remove blood from the aorta to take it to the hemoconcentrator decreases the effective cardiac output. This would be especially true in patients who have decreased cardiac reserves, such as after repair of total anomalous pulmonary venous return in which the stroke volume and the cardiac output are marginal anyway. To decrease the cardiac output any further may be detrimental. Not only is there a decrease in the cardiac output, but arteriovenous MUF may volume overload the patient with its attendant left-to-right shunt. Thus we have not performed arteriovenous MUF for these purposes.
In another group of patients, specifically those who have a systemic–pulmonary artery shunt in which there is already a left-to-right shunt, adding any further left-to-right shunting may further decrease the perfusion and increase the volume load to the ventricle. I emphasize that these are theoretical considerations, and these are the reasons that we have stayed away from arteriovenous MUF.
Dr Gaynor, you asked whether we looked at the hemodynamics. We have not. I have enjoyed looking at your work, particularly the pressure/volume relationships with arteriovenous MUF. We have not duplicated those studies.
Zero balance hemoconcentration is effective. This is merely an aggressive way of removing almost all of the priming volume. Our goal is to remove all of the priming volume, but in practice it is actually hard to remove the entire priming volume because of volume concerns.
We see an advantage to both conventional and MUF. Our current practice is to perform CUF during the rewarming phase of CPB, remove as much ultrafiltrate as possible during that time, and then use venovenous MUF to remove the remaining portion of the priming volume. In our daily practice, we use a combination of conventional and venovenous MUF.
I agree that the weakness in this study is that it does not compare arteriovenous with venovenous ultrafiltration, but we are not familiar with arteriovenous ultrafiltration, and for those concerns that I have mentioned, we have not had an opportunity to try it.
Dr Ko Bando (Indianapolis, Ind). We are also interested in using venovenous MUF for patients with complex congenital heart disease. From your diagram, if I am correct, it seems to me that you are returning nonoxygenated blood to the patient. If that is the case, there is a concern that you may increase the pulmonary vascular reactivity and thus increase the incidence of pulmonary hypertensive crises after venovenous MUF. Did you see any postoperative pulmonary hypertensive crises among patients with MUF? Were there any differences in the postoperative pulmonary arterial pressure among the 3 groups?
Dr Hennein. We take it from the IVC cannula, which is desaturated blood, hemoconcentrated, and then return it to the same cardiac chamber. Therefore it is deoxygenated blood, and it is slightly more deoxygenated than when it first came in to be sure, but it is not that much more deoxygenated. We have had no problems with desaturation.
But Dr Bando does bring up a good point, that you can hemoconcentrate by taking from the venous side, hemoconcentrating, and then returning to the arterial side; this has been done in some centers, and it is particularly for that concern that this causes a right-to-left shunt and would lead in cyanosis. We do not use this technique.
The other standard way is the arteriovenous way, which, as Dr Gaynor and others of our colleagues have done, is to take blood from the arterial cannula, reverse flow into the ultrafilter, and then return it to the systemic side. This is the left-to-right shunting about which we have concerns.
In general, we believe that venovenous MUF offers no hemodynamic disadvantages in that it causes neither a left-to-right nor a right-to-left shunt. It is a very hemodynamically stable system, and it is highly efficient.
Dr Richard A. Jonas (Boston, Mass). I noticed that you were using a mean prime volume of about 1300 mL and that you extended hemodilution down to a hematocrit value of 22% and also that your control patients spent about 5 days in the ICU.
Why hemodilute to such an extent? Why add all that water at the beginning? With today's modern circuits, it is possible to reduce priming volumes, and our data and other people's experiences suggest that you do not need to add all that water in the first place. Is the hemodilution you are using simply creating a problem that you are subsequently having to treat with ultrafiltration?
Is it possible to overdo this? Did you look at renal function, creatinine, and blood urea nitrogen? Do you think you can prevent the usual diuresis that occurs after operation because you have over-hemoconcentrated? Even though it might not be important in a patient who has excellent hemodynamics, excessive hemoconcentration in a small sick neonate might create problems with a higher risk of renal failure.
Dr Hennein. The amount of the priming volume appears to be of concern because of our patient mix. This was a consecutive series of 38 patients, and they were treated sequentially. The manuscript describes several operations in neonates, for example, 2 or 3 switches, a couple of total anomalous pulmonary venous returns, and other operations; in those patients we obviously use a much lower priming volume. However, the great majority of the patients who came in during this period were larger patients, so their priming volume was high.
It is our goal also to continually decrease the amount of the priming volume, and we could continue to do a better job. Although in the neonates we do have priming volumes in the 500- to 600-mL range, it would be nice to get lower. However, that is the lower limits of what we have been doing at the present time.
Regarding whether we are overdoing the hemoconcentration, it is possible but not likely. We really had no major complications in this group of patients. They did reasonably well. The ICU stay was a little bit longer than I would have anticipated in the control group. It was directly related to the amount of total body water that was gained, so I would say that we may not have overdone it, but there is a distinct benefit to patients who had ultrafiltration. There were no problems with renal failure or renal shutdown afterward.
Ultrafiltration is really limited by 2 factors. One is that the goal of ultrafiltration is to remove the priming volume. The other factor is that often one simply runs out of volume in the CPB unit and must stop the MUF process.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionConclusionsAppendix: Discussion References |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Charette, Y. Hirata, A. Bograd, L. Mongero, J. Chen, J. Quaegebeur, and R. Mosca 180 ml and less: Cardiopulmonary bypass techniques to minimize hemodilution for neonates and small infants Perfusion, September 1, 2007; 22(5): 327 - 331. [Abstract] [PDF]
|
|
|
|
 |
|
 G. D. Williams, C. Ramamoorthy, L. Chu, G. B. Hammer, K. Kamra, M. G. Boltz, K. Pentcheva, J. P. McCarthy, and V. M. Reddy Modified and conventional ultrafiltration during pediatric cardiac surgery: Clinical outcomes compared J. Thorac. Cardiovasc. Surg., December 1, 2006; 132(6): 1291 - 1298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Li, A. Hoschtitzky, M. L. Allen, M. J. Elliott, and A. N. Redington An Analysis of Oxygen Consumption and Oxygen Delivery in Euthermic Infants After Cardiopulmonary Bypass With Modified Ultrafiltration Ann. Thorac. Surg., October 1, 2004; 78(4): 1389 - 1396. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S Chew, V. Brix-Christensen, H. B Ravn, I. Brandslund, E. Ditlevsen, J. Pedersen, K. Hjortholm, O. K Hansen, E. Tunnesen, and V. E Hjortdal Effect of modified ultrafiltration on the inflammatory response in paediatric open-heart surgery: a prospective, randomized study Perfusion, September 1, 2002; 17(5): 327 - 333. [Abstract] [PDF]
|
|
|
|
|
|
H. A Hennein Venovenous modified ultrafiltration after cardiopulmonary bypass in children: A prospective randomized study J. Thorac. Cardiovasc. Surg., March 1, 2000; 119(3): 630 - 630. [Full Text] [PDF]
|
|
|
|
 |
|
 J. Grunenfelder, G. Zund, A. Schoeberlein, F. E. Maly, U. Schurr, S. Guntli, K. Fischer, and M. Turina Modified ultrafiltration lowers adhesion molecule and cytokine levels after cardiopulmonary bypass without clinical relevance in adults Eur. J. Cardiothorac. Surg., January 1, 2000; 17(1): 77 - 83. [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 |
