Perfusion
http://prf.sagepub.com/
Clinical outcome and blood transfusion after infant cardiac surgery with a routine use of conventional ultrafiltration HD Golab, J Kissler, PL de Jong, PC van de Woestijne, JJM Takkenberg and AJJC Bogers Perfusion published online 13 August 2014 DOI: 10.1177/0267659114546946 The online version of this article can be found at: http://prf.sagepub.com/content/early/2014/08/13/0267659114546946
Published by: http://www.sagepublications.com
Additional services and information for Perfusion can be found at: Email Alerts: http://prf.sagepub.com/cgi/alerts Subscriptions: http://prf.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Aug 13, 2014 What is This?
Downloaded from prf.sagepub.com at GEORGIAN COURT UNIV on November 14, 2014
46946 research-article2014
PRF0010.1177/0267659114546946PerfusionGolab et al.
Original paper
Clinical outcome and blood transfusion after infant cardiac surgery with a routine use of conventional ultrafiltration
Perfusion 1–9 © The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0267659114546946 prf.sagepub.com
HD Golab,1 J Kissler,2 PL de Jong,1 PC van de Woestijne,1 JJM Takkenberg1 and AJJC Bogers1
Abstract Objective. Priming-related hemodilution is the culprit behind excessive body water accumulation, postoperative coagulopathy and enhanced blood transfusion in infant cardiac surgery patients. In this retrospective, observational study, clinical data were analyzed to assess the effect of conventional ultrafiltration on allogenic blood transfusion and patient clinical outcome. Methods. All infants with a bodyweight up to 10 kg who underwent consequent cardiac surgery in 2011 and 2012 were eligible for the audit. Seventy patients, operated in accordance with existing pediatric protocol, enrolled in the control group. The study group consisted of 55 patients who were operated employing conventional ultrafiltration during bypass and recently adjusted hematocrit targets. The following variables were primarily investigated: hematocrit and colloid osmotic pressure value, total volume of blood products transfused and duration of postoperative mechanical ventilation. Secondary outcome measures were: postoperative urine production, postoperative blood loss, length of stay at the intensive care unit and hospital stay. Results. There were no significant differences between the groups in relation to demographics or hematological and cardiopulmonary bypass data. The ultrafiltration volume removed from circulation during bypass in the study group was 171 ± 99 ml. No significant difference between the groups was found with regard to the total allogenic blood transfusion (study group 216 ± 92 ml versus control group 191 ±93 ml; p = 0.136). All recorded clinical end points, duration of mechanical ventilation, duration of chest tube in situ, stay in ICU and stay in hospital, were similar between the groups. Conclusions. Routine use of conventional ultrafiltration during the cardiac surgery for patients with a bodyweight less than 10 kg was a safe technique that allowed us to achieve higher hematocrit levels at the end of the operation without additional transfusions of allogenic blood. On the other hand, ultrafiltration did not improve the clinical end points. Keywords infant cardiac surgery; cardiopulmonary bypass; conventional ultrafiltration; blood transfusion; clinical outcome
Introduction Despite the technological advances in cardiopulmonary bypass (CPB) circuits for pediatric patients, smaller children who undergo heart surgery with lengthy CPB times are at risk for priming-related hemodilution,1,2 excessive body water accumulation,3,4 postoperative coagulopathy and increased blood transfusion.5,6 To counteract those adverse effects of CPB, different ultrafiltration techniques were introduced. Primarily, ultrafiltration was merely used to remove excess body water. Later on, the possibility of the removal of pro-inlammatory mediators was explored.7,8 As conventional ultrafiltration (CUF) carried out during CPB has only limited filtration efficiency,
1Department
of Cardiothoracic Surgery, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands 2Department of Anaesthesiology, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands Corresponding author: HD Golab Erasmus MC Postbus 2040 3000 CE Rotterdam Bd 199 The Netherlands. Email: [email protected]
Downloaded from prf.sagepub.com at GEORGIAN COURT UNIV on November 14, 2014
2
Perfusion
modified ultrafiltration (MUF), performed immediately after the cessation of CPB, was developed.9 Both ultrafiltration techniques are, nowadays, broadly used in pediatric cardiac surgery and numerous clinical studies have found both ultrafiltration techniques to be beneficial in relation to unfiltrated CPB.10–12 There are some important differences between CUF and MUF. CUF does not prolong the procedure time and allows achieving only moderate amounts of hemoconcentration, but is more efficient in the removal of inflammatory cytokines. MUF requires additional time (15 to 20 min) after the cessation of CPB, increases the complexity of the procedure and may even transiently misbalance patient hemodynamics.13 On the other hand, MUF provides more effective hemoconcentration in the immediate post-CPB period.14,15 Both ultrafiltration techniques are considered safe and there were no significant differences found in the postoperative outcome parameters comparing both techniques.11,12 Therefore, the relative risk and institutional objective may influence the choice of ultrafiltration technique. In our institution, the CPB circuit for infant patients was miniaturized to achieve a priming volume of 230 ml for children with a bodyweight up to 10 kg.16 We adjusted the management of colloid osmotic pressure (COP) during CPB17 and introduced an autologous transfusion device (ATD) to collect and process shed blood and the residual CPB circuit volume.18,19 In 2012, to optimize patients’ clinical outcome, CUF was introduced together with an adjusted protocol for hematocrit targets during and at the end of CPB. The aim of this retrospective study was to compare clinical outcome and allogenic blood product transfusion between two cohorts of infants who underwent elective cardiac surgery in our institution. The first cohort (control group) was operated in 2011 in accordance with the existing pediatric protocol. On the assumption that ultrafiltration is beneficial, the second cohort (study group), operated in 2012, received CPB with CUF and newly established hematocrit targets. We hypothesized that the CUF group would have a better clinical outcome than the control group and that the amount of allogenic blood transfusion would be diminished.
Materials and Methods This retrospective primary data audit was performed in accordance with the Institutional Data Protection Policy and received approval of the Medical Ethical Commission of Erasmus MC (MEC 2013-081).
Study Design All consecutive infants with a bodyweight up to 10 kg who underwent cardiac surgery between January 2011
and January 2013 were eligible for the audit. Exclusion criteria were: pre-existing non-cardiac disease that could jeopardize postoperative recovery and direct postoperative hemodynamic instability that required operative revision or artificial circulatory support. Patients in the control group were operated in the calendar year 2011 in accordance with the existing pediatric protocol. Patients in the study group were those operated in calendar year 2012, conforming to our 2011 pediatric protocol, with supplemental CUF during CPB and newly adjusted hematocrit targets depending on the correction type and patient bodyweight. In the study group, cerebral oxygenation was additionally assessed by near infrared spectroscopy (NIRS) to control adequacy of perfusion.20,22 The same surgical team (3 surgeons, 3 anesthesiologists, 10 perfusionists, all working within the protocol) performed all the operations. The following variables were primarily investigated: (1) hematocrit and colloid osmotic pressure value at the end of CPB, (2) total volume of blood products transfused and (3) duration of postoperative mechanical ventilation. Secondary outcome measures were: postoperative urine production, postoperative blood loss, length of stay in the intensive care unit (ICU) and postoperative hospital stay. We monitored the incidence of postoperative complications and mortality in both cohorts. Blood analysis included hemoglobin (Hb) concentration, hematocrit (Hct), platelet count (Plt) and lactate concentration (Lac). Laboratory data were obtained starting one day before the operation until 24 hours after surgery. The volume of crystalloids, colloids and blood products transfused in the operation room (OR) and at the ICU were recorded.
Cardiopulmonary bypass, anesthesia and anticoagulation All patients received standard anesthesia as described previously.23 Infant CPB circuits consisted of a Capiox Baby Fx 05 hollow-fiber oxygenator with a hard-shell reservoir and integrated arterial filter (Terumo, Tokyo, Japan) and a roller pump. Tubing internal diameter of the circuit and the pump boot was ¼ inch. The circuits were phosphorylcholine coated (Sorin Group, Mirandola, Italy). Priming contained homologous red blood cells (RBC), fresh-frozen plasma (FFP) and Gelofusine (Braun, Melsungen, Germany). The amount of RBC in the prime in the control group was calculated to achieve a hematocrit of 28% at the start of CPB for all patients, irrespective of bodyweight and correction type. The target at the end of CPB in the control group was established equally at 28%. Hematocrit targets in the study group are presented in the Table 1.
Downloaded from prf.sagepub.com at GEORGIAN COURT UNIV on November 14, 2014
3
Golab et al. Table 1. Hematocrit targets of the study group. Patient body weight
< 5 kg
5 – 10 kg
Target at the start of CPB Univentricular correction Biventricular correction
28% 25%
25% 23%
Target at the end of CPB Univentricular correction Biventricular correction
35% 30%
33% 28%
CPB: cardiopulmonary bypass.
The ratio of FFP to Gelofusine in the prime was 1 to 1. The prime was always completed with an amount of human albumin 20% solution (Table 3), 0.5 g/kg mannitol, 2-5 ml sodium bicarbonate 8.4% and 4.2 IU heparin/ml priming volume. Activated clotting time was monitored during the bypass and kept above 480 sec. Post-bypass protamine sulphate was administrated to neutralize heparin. Non-pulsatile CPB with mild hypothermia was performed with blood flow rates between 1.8 L/min/m2 to 3.2 L/min/m2 to maintain venous blood oxygen saturation (SvO2) above 65% and the mean arterial pressure between 40 to 60 mmHg. During CPB, α-Stat regulation was used unless deep hypothermia was required for which a pH-Stat regulation during cooling was applied. Adequacy of perfusion was always assessed by a continuous “on-line” monitoring of arterial blood gas parameters, the SvO2 and haematocrit, measured by a CDI 500 (Terumo, Tokyo, Japan). At the start and at the end of CPB, laboratory blood gas and hematological analyses were executed. In the study group, cerebral oximetry was performed by the Invos Oximeter (Somanetics, Troy, MI, USA). If there were difficulties in maintaining the SvO2 above 65% and cerebral oximetry above 50% despite increasing the pump flow and oxygen concentration during bypass in presence of an adequate partial pressure of carbon dioxide in the arterial blood, allogeneic blood transfusion was administrated. Myocardial protection was achieved with cold crystalloid cardioplegia that was preferably aspirated into an autotransfusion device (ATD) (Electa, Sorin, Milan, Italy). In the study group, CUF was performed with a Jostra BC 20 plus (Maquet, Hirrlingen, Germany) polyacrylsulfon hemofilter. The hemofilter set was pre-primed with 50 ml NaCl 0.9% and connected between the oxygenator inlet line and the venous reservoir. No pump was used. During ultrafiltration, blood flowed from the high pressure of the oxygenator inlet line to the low pressure of the venous reservoir. To increase the transmembrane pressure, the hemofilter outlet tubing was partly clamped and that also decreased the “stolen” blood flow.24 Ultrafiltration was performed intermittently throughout CPB in relation to the availability of circulating volume and the hematocrit target at the end
of CPB. Perioperative blood loss and residual volume of the CPB circuit were collected and processed by the ATD. Administration of RBC, FFP, crystalloids or colloids during CPB was at the discretion of perfusionist with regard to the target values for hematocrit and COP and the system working volumes. The target COP was ⩾18 mmHg. Before and after CPB, fluids and blood product transfusion was based upon the patient clinical status. The ATD product, if available, was always considered first-line blood replacement therapy.
Statistical analysis Continuous variables are presented as a mean ± standard deviation (SD); categorical data are presented as proportions. Continuous independent data were compared with an unpaired t test (in the case of normally distributed data) or the Mann – Whitney test (in the case of non-normally distributed data). Categorical data were compared with the Chi-square test. Repeated measures of continuous variables were compared using a paired t-test. Univariate binary logistic regression analysis (method enter) was used to evaluate predictors associated with an increased risk for low hematocrit level at the end of CPB and for the abolishing of CUF protocol. All tests were two-sided and a p-value less than 0.05 were considered statistically significant. All statistical analyses were performed using SPSS 17.0 statistical software (SPSS, Chicago, IL).
Results In the year 2011, we operated on 77 patients with congenital heart disease and a bodyweight less than 10 kg. Four infants were excluded from the study due to the exclusion criteria, together with an additional three patients who received CUF because of hypervolemic hemodilution. The remaining 70 patients were enrolled in the control group. The study group consisted of 55 patients out of 76 eligible infants operated on in 2012. There were five exclusions according to the exclusion criteria and, in 16 patients, CUF was not applied. This group of 16 patients consisted of slightly younger (4.4 ± 5.5 months, p = 0.15) and significantly smaller (4.7 ± 2.0 kg, p = 0.03) infants. The attending perfusionists made the decision not to apply CUF, taking circulating volume and actual hematocrit at the start of CPB into account, together with assumed results during perfusion. One of those patients was subjected to a procedure with deep hypothermic circulatory arrest with anterograde cerebral perfusion and there were two hospital deaths in this group.
Downloaded from prf.sagepub.com at GEORGIAN COURT UNIV on November 14, 2014
4
Perfusion
Table 2. Demographics of the study population. Variable
Male/female Age (month) Weight (kg) BSA (dm2) CPB time (min) AoX time (min) DHCA/ACP Hospital mortality Rethoracotomy 24 h RACHS -1 1 2 3 4 5 6
Control group
Study group
N = 70
N = 55
36/34 5.9 ± 5.0 5.8 ± 2.0 33 ± 8 103 ± 58 61 ± 47 7 2 1
30/25 6.5 ± 4.9 6.0 ± 2.1 33 ± 9 107 ± 59 65 ± 38 3 1 2
8 24 26 12 0 0
5 30 13 6 0 1
P value
0.52 0.60 0.63 0.70 0.61
All values are presented as a mean ± SD; AoX: aorta occlusion; BSA: body surface area; CPB: cardiopulmonary bypass; DHCA/ACP: deep hypothermic circulatory arrest with anterograde cerebral perfusion; RACHS: risk adjustment in surgery for congenital heart disease.
Table 3. Priming composition and volume transfusion during CPB. Variable
Control group Study group P value
N = 70
Priming volume (ml) 260 ± 28 RBC priming (ml) 93 ± 49 FFP priming (ml) 57 ± 26 Gelofusine priming (ml) 58 ± 26 Albumin priming (ml) 34 ± 7 RBC CPB (ml) 49 ± 49 FFP CPB (ml) 19 ± 26 Gelofusine CPB (ml) 26 ± 47 Albumin CPB (ml) 4±7 Cardioplegia CPB (ml) 45 ± 45
N = 55 258 ± 24 57 ±51 73 ± 26 70 ± 26 32 ± 6 118 ± 65 28 ± 32 31 ± 37 2±4 59 ± 62
0.603 0.000 0.001 0.016 0.281 0.000 0.070 0.562 0.147 0.140
All values are presented as a mean ± SD. Cardioplegia CPB: cardioplegia introduced into the circulation; CPB: cardiopulmonary bypass; FFP: fresh frozen plasma; RBC: red blood cells.
Demographic characteristics and clinical data of the control and the study group are presented in Table 2. There were no significant differences between the groups in relation to the duration of the CPB and aorta occlusion. No complications were identified related to the ultrafiltration technique. The total priming volume of the CPB circuit did not differ between the study and control groups nor the ratio between the Gelofusine and FFP used in the prime (study group: 1.04 versus control group: 0.98, p = 0.89).
On the other hand, the amount of RBC in the prime was significantly different between the groups as a consequence of different hematocrit targets at the start of CPB (Table 3) and that, additionally, resulted in significant differences in the volume of Gelofusine and FFP. In both groups, no crystalloids was used as the volume therapy during CPB. The mean ultrafiltration volume removed in the study group was 171 ± 99 ml (min 35 ml, max 480 ml, median 150 ml). No significant differences between the groups were found with regard to the total amount of allogeneic blood transfusion in the study period (study group 216 ± 92 ml versus control group 191 ± 93 ml; p = 0.136, Figure 1). During the ICU stay, only 2 out of 55 patients in the study group received allogeneic blood whereas 52 out of 70 patients in the control group were transfused with homologous RBC. Post-CPB transfusion of FFP, platelet concentrate and autologous blood product did not differ significantly between the groups; the study group received a total transfusion of 236 ± 204 ml versus 211 ± 198 ml (p = 0.26) in the control group. The groups did not differ significantly in respect to the preoperative values of Hb, Hct, Plt (Figure 2 a, b, c) and Lac (control group 1.4 ± 1.0 mmol/l versus study group 1.2 ± 0.5 mmol/l, p = 0.35). At the end of CPB, the study group had a significantly higher hematocrit than the control group and this significant difference in favour of the study group remained till the end of the operation. However, after 24 hours, the groups did not differ in respect to the hematocrit value. Plasma lactate concentration at the start of CPB (control group 1.6 ± 0.7 mmol/l versus study group 1.7 ± 0.8 mmol/l, p = 0.30) and at the end of operation (control group 1.6 ± 1.0 mmol/l versus study group 1.7 ± 1.1 mmol/l, p = 0.49) did not show significant differences between the groups. The COP in the study group did not change significantly during CPB (19.8 ± 2.1 mmHg at the start and 19.3 ± 2.3 mmHg at the end of bypass, p = 0.76). The COP of the control group diminished significantly during CPB from 19.4 ± 1.9 mmHg at the start to 17.6 ± 2.1 mmHg at the end of CPB, p = 0.003). Therefore, at the end of bypass, the COP of the study group was significantly higher compared to the control group (19.3 ± 2.3 mmHg versus 17.6 ± 2.1 mmHg, p
http://prf.sagepub.com/
Clinical outcome and blood transfusion after infant cardiac surgery with a routine use of conventional ultrafiltration HD Golab, J Kissler, PL de Jong, PC van de Woestijne, JJM Takkenberg and AJJC Bogers Perfusion published online 13 August 2014 DOI: 10.1177/0267659114546946 The online version of this article can be found at: http://prf.sagepub.com/content/early/2014/08/13/0267659114546946
Published by: http://www.sagepublications.com
Additional services and information for Perfusion can be found at: Email Alerts: http://prf.sagepub.com/cgi/alerts Subscriptions: http://prf.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Aug 13, 2014 What is This?
Downloaded from prf.sagepub.com at GEORGIAN COURT UNIV on November 14, 2014
46946 research-article2014
PRF0010.1177/0267659114546946PerfusionGolab et al.
Original paper
Clinical outcome and blood transfusion after infant cardiac surgery with a routine use of conventional ultrafiltration
Perfusion 1–9 © The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0267659114546946 prf.sagepub.com
HD Golab,1 J Kissler,2 PL de Jong,1 PC van de Woestijne,1 JJM Takkenberg1 and AJJC Bogers1
Abstract Objective. Priming-related hemodilution is the culprit behind excessive body water accumulation, postoperative coagulopathy and enhanced blood transfusion in infant cardiac surgery patients. In this retrospective, observational study, clinical data were analyzed to assess the effect of conventional ultrafiltration on allogenic blood transfusion and patient clinical outcome. Methods. All infants with a bodyweight up to 10 kg who underwent consequent cardiac surgery in 2011 and 2012 were eligible for the audit. Seventy patients, operated in accordance with existing pediatric protocol, enrolled in the control group. The study group consisted of 55 patients who were operated employing conventional ultrafiltration during bypass and recently adjusted hematocrit targets. The following variables were primarily investigated: hematocrit and colloid osmotic pressure value, total volume of blood products transfused and duration of postoperative mechanical ventilation. Secondary outcome measures were: postoperative urine production, postoperative blood loss, length of stay at the intensive care unit and hospital stay. Results. There were no significant differences between the groups in relation to demographics or hematological and cardiopulmonary bypass data. The ultrafiltration volume removed from circulation during bypass in the study group was 171 ± 99 ml. No significant difference between the groups was found with regard to the total allogenic blood transfusion (study group 216 ± 92 ml versus control group 191 ±93 ml; p = 0.136). All recorded clinical end points, duration of mechanical ventilation, duration of chest tube in situ, stay in ICU and stay in hospital, were similar between the groups. Conclusions. Routine use of conventional ultrafiltration during the cardiac surgery for patients with a bodyweight less than 10 kg was a safe technique that allowed us to achieve higher hematocrit levels at the end of the operation without additional transfusions of allogenic blood. On the other hand, ultrafiltration did not improve the clinical end points. Keywords infant cardiac surgery; cardiopulmonary bypass; conventional ultrafiltration; blood transfusion; clinical outcome
Introduction Despite the technological advances in cardiopulmonary bypass (CPB) circuits for pediatric patients, smaller children who undergo heart surgery with lengthy CPB times are at risk for priming-related hemodilution,1,2 excessive body water accumulation,3,4 postoperative coagulopathy and increased blood transfusion.5,6 To counteract those adverse effects of CPB, different ultrafiltration techniques were introduced. Primarily, ultrafiltration was merely used to remove excess body water. Later on, the possibility of the removal of pro-inlammatory mediators was explored.7,8 As conventional ultrafiltration (CUF) carried out during CPB has only limited filtration efficiency,
1Department
of Cardiothoracic Surgery, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands 2Department of Anaesthesiology, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands Corresponding author: HD Golab Erasmus MC Postbus 2040 3000 CE Rotterdam Bd 199 The Netherlands. Email: [email protected]
Downloaded from prf.sagepub.com at GEORGIAN COURT UNIV on November 14, 2014
2
Perfusion
modified ultrafiltration (MUF), performed immediately after the cessation of CPB, was developed.9 Both ultrafiltration techniques are, nowadays, broadly used in pediatric cardiac surgery and numerous clinical studies have found both ultrafiltration techniques to be beneficial in relation to unfiltrated CPB.10–12 There are some important differences between CUF and MUF. CUF does not prolong the procedure time and allows achieving only moderate amounts of hemoconcentration, but is more efficient in the removal of inflammatory cytokines. MUF requires additional time (15 to 20 min) after the cessation of CPB, increases the complexity of the procedure and may even transiently misbalance patient hemodynamics.13 On the other hand, MUF provides more effective hemoconcentration in the immediate post-CPB period.14,15 Both ultrafiltration techniques are considered safe and there were no significant differences found in the postoperative outcome parameters comparing both techniques.11,12 Therefore, the relative risk and institutional objective may influence the choice of ultrafiltration technique. In our institution, the CPB circuit for infant patients was miniaturized to achieve a priming volume of 230 ml for children with a bodyweight up to 10 kg.16 We adjusted the management of colloid osmotic pressure (COP) during CPB17 and introduced an autologous transfusion device (ATD) to collect and process shed blood and the residual CPB circuit volume.18,19 In 2012, to optimize patients’ clinical outcome, CUF was introduced together with an adjusted protocol for hematocrit targets during and at the end of CPB. The aim of this retrospective study was to compare clinical outcome and allogenic blood product transfusion between two cohorts of infants who underwent elective cardiac surgery in our institution. The first cohort (control group) was operated in 2011 in accordance with the existing pediatric protocol. On the assumption that ultrafiltration is beneficial, the second cohort (study group), operated in 2012, received CPB with CUF and newly established hematocrit targets. We hypothesized that the CUF group would have a better clinical outcome than the control group and that the amount of allogenic blood transfusion would be diminished.
Materials and Methods This retrospective primary data audit was performed in accordance with the Institutional Data Protection Policy and received approval of the Medical Ethical Commission of Erasmus MC (MEC 2013-081).
Study Design All consecutive infants with a bodyweight up to 10 kg who underwent cardiac surgery between January 2011
and January 2013 were eligible for the audit. Exclusion criteria were: pre-existing non-cardiac disease that could jeopardize postoperative recovery and direct postoperative hemodynamic instability that required operative revision or artificial circulatory support. Patients in the control group were operated in the calendar year 2011 in accordance with the existing pediatric protocol. Patients in the study group were those operated in calendar year 2012, conforming to our 2011 pediatric protocol, with supplemental CUF during CPB and newly adjusted hematocrit targets depending on the correction type and patient bodyweight. In the study group, cerebral oxygenation was additionally assessed by near infrared spectroscopy (NIRS) to control adequacy of perfusion.20,22 The same surgical team (3 surgeons, 3 anesthesiologists, 10 perfusionists, all working within the protocol) performed all the operations. The following variables were primarily investigated: (1) hematocrit and colloid osmotic pressure value at the end of CPB, (2) total volume of blood products transfused and (3) duration of postoperative mechanical ventilation. Secondary outcome measures were: postoperative urine production, postoperative blood loss, length of stay in the intensive care unit (ICU) and postoperative hospital stay. We monitored the incidence of postoperative complications and mortality in both cohorts. Blood analysis included hemoglobin (Hb) concentration, hematocrit (Hct), platelet count (Plt) and lactate concentration (Lac). Laboratory data were obtained starting one day before the operation until 24 hours after surgery. The volume of crystalloids, colloids and blood products transfused in the operation room (OR) and at the ICU were recorded.
Cardiopulmonary bypass, anesthesia and anticoagulation All patients received standard anesthesia as described previously.23 Infant CPB circuits consisted of a Capiox Baby Fx 05 hollow-fiber oxygenator with a hard-shell reservoir and integrated arterial filter (Terumo, Tokyo, Japan) and a roller pump. Tubing internal diameter of the circuit and the pump boot was ¼ inch. The circuits were phosphorylcholine coated (Sorin Group, Mirandola, Italy). Priming contained homologous red blood cells (RBC), fresh-frozen plasma (FFP) and Gelofusine (Braun, Melsungen, Germany). The amount of RBC in the prime in the control group was calculated to achieve a hematocrit of 28% at the start of CPB for all patients, irrespective of bodyweight and correction type. The target at the end of CPB in the control group was established equally at 28%. Hematocrit targets in the study group are presented in the Table 1.
Downloaded from prf.sagepub.com at GEORGIAN COURT UNIV on November 14, 2014
3
Golab et al. Table 1. Hematocrit targets of the study group. Patient body weight
< 5 kg
5 – 10 kg
Target at the start of CPB Univentricular correction Biventricular correction
28% 25%
25% 23%
Target at the end of CPB Univentricular correction Biventricular correction
35% 30%
33% 28%
CPB: cardiopulmonary bypass.
The ratio of FFP to Gelofusine in the prime was 1 to 1. The prime was always completed with an amount of human albumin 20% solution (Table 3), 0.5 g/kg mannitol, 2-5 ml sodium bicarbonate 8.4% and 4.2 IU heparin/ml priming volume. Activated clotting time was monitored during the bypass and kept above 480 sec. Post-bypass protamine sulphate was administrated to neutralize heparin. Non-pulsatile CPB with mild hypothermia was performed with blood flow rates between 1.8 L/min/m2 to 3.2 L/min/m2 to maintain venous blood oxygen saturation (SvO2) above 65% and the mean arterial pressure between 40 to 60 mmHg. During CPB, α-Stat regulation was used unless deep hypothermia was required for which a pH-Stat regulation during cooling was applied. Adequacy of perfusion was always assessed by a continuous “on-line” monitoring of arterial blood gas parameters, the SvO2 and haematocrit, measured by a CDI 500 (Terumo, Tokyo, Japan). At the start and at the end of CPB, laboratory blood gas and hematological analyses were executed. In the study group, cerebral oximetry was performed by the Invos Oximeter (Somanetics, Troy, MI, USA). If there were difficulties in maintaining the SvO2 above 65% and cerebral oximetry above 50% despite increasing the pump flow and oxygen concentration during bypass in presence of an adequate partial pressure of carbon dioxide in the arterial blood, allogeneic blood transfusion was administrated. Myocardial protection was achieved with cold crystalloid cardioplegia that was preferably aspirated into an autotransfusion device (ATD) (Electa, Sorin, Milan, Italy). In the study group, CUF was performed with a Jostra BC 20 plus (Maquet, Hirrlingen, Germany) polyacrylsulfon hemofilter. The hemofilter set was pre-primed with 50 ml NaCl 0.9% and connected between the oxygenator inlet line and the venous reservoir. No pump was used. During ultrafiltration, blood flowed from the high pressure of the oxygenator inlet line to the low pressure of the venous reservoir. To increase the transmembrane pressure, the hemofilter outlet tubing was partly clamped and that also decreased the “stolen” blood flow.24 Ultrafiltration was performed intermittently throughout CPB in relation to the availability of circulating volume and the hematocrit target at the end
of CPB. Perioperative blood loss and residual volume of the CPB circuit were collected and processed by the ATD. Administration of RBC, FFP, crystalloids or colloids during CPB was at the discretion of perfusionist with regard to the target values for hematocrit and COP and the system working volumes. The target COP was ⩾18 mmHg. Before and after CPB, fluids and blood product transfusion was based upon the patient clinical status. The ATD product, if available, was always considered first-line blood replacement therapy.
Statistical analysis Continuous variables are presented as a mean ± standard deviation (SD); categorical data are presented as proportions. Continuous independent data were compared with an unpaired t test (in the case of normally distributed data) or the Mann – Whitney test (in the case of non-normally distributed data). Categorical data were compared with the Chi-square test. Repeated measures of continuous variables were compared using a paired t-test. Univariate binary logistic regression analysis (method enter) was used to evaluate predictors associated with an increased risk for low hematocrit level at the end of CPB and for the abolishing of CUF protocol. All tests were two-sided and a p-value less than 0.05 were considered statistically significant. All statistical analyses were performed using SPSS 17.0 statistical software (SPSS, Chicago, IL).
Results In the year 2011, we operated on 77 patients with congenital heart disease and a bodyweight less than 10 kg. Four infants were excluded from the study due to the exclusion criteria, together with an additional three patients who received CUF because of hypervolemic hemodilution. The remaining 70 patients were enrolled in the control group. The study group consisted of 55 patients out of 76 eligible infants operated on in 2012. There were five exclusions according to the exclusion criteria and, in 16 patients, CUF was not applied. This group of 16 patients consisted of slightly younger (4.4 ± 5.5 months, p = 0.15) and significantly smaller (4.7 ± 2.0 kg, p = 0.03) infants. The attending perfusionists made the decision not to apply CUF, taking circulating volume and actual hematocrit at the start of CPB into account, together with assumed results during perfusion. One of those patients was subjected to a procedure with deep hypothermic circulatory arrest with anterograde cerebral perfusion and there were two hospital deaths in this group.
Downloaded from prf.sagepub.com at GEORGIAN COURT UNIV on November 14, 2014
4
Perfusion
Table 2. Demographics of the study population. Variable
Male/female Age (month) Weight (kg) BSA (dm2) CPB time (min) AoX time (min) DHCA/ACP Hospital mortality Rethoracotomy 24 h RACHS -1 1 2 3 4 5 6
Control group
Study group
N = 70
N = 55
36/34 5.9 ± 5.0 5.8 ± 2.0 33 ± 8 103 ± 58 61 ± 47 7 2 1
30/25 6.5 ± 4.9 6.0 ± 2.1 33 ± 9 107 ± 59 65 ± 38 3 1 2
8 24 26 12 0 0
5 30 13 6 0 1
P value
0.52 0.60 0.63 0.70 0.61
All values are presented as a mean ± SD; AoX: aorta occlusion; BSA: body surface area; CPB: cardiopulmonary bypass; DHCA/ACP: deep hypothermic circulatory arrest with anterograde cerebral perfusion; RACHS: risk adjustment in surgery for congenital heart disease.
Table 3. Priming composition and volume transfusion during CPB. Variable
Control group Study group P value
N = 70
Priming volume (ml) 260 ± 28 RBC priming (ml) 93 ± 49 FFP priming (ml) 57 ± 26 Gelofusine priming (ml) 58 ± 26 Albumin priming (ml) 34 ± 7 RBC CPB (ml) 49 ± 49 FFP CPB (ml) 19 ± 26 Gelofusine CPB (ml) 26 ± 47 Albumin CPB (ml) 4±7 Cardioplegia CPB (ml) 45 ± 45
N = 55 258 ± 24 57 ±51 73 ± 26 70 ± 26 32 ± 6 118 ± 65 28 ± 32 31 ± 37 2±4 59 ± 62
0.603 0.000 0.001 0.016 0.281 0.000 0.070 0.562 0.147 0.140
All values are presented as a mean ± SD. Cardioplegia CPB: cardioplegia introduced into the circulation; CPB: cardiopulmonary bypass; FFP: fresh frozen plasma; RBC: red blood cells.
Demographic characteristics and clinical data of the control and the study group are presented in Table 2. There were no significant differences between the groups in relation to the duration of the CPB and aorta occlusion. No complications were identified related to the ultrafiltration technique. The total priming volume of the CPB circuit did not differ between the study and control groups nor the ratio between the Gelofusine and FFP used in the prime (study group: 1.04 versus control group: 0.98, p = 0.89).
On the other hand, the amount of RBC in the prime was significantly different between the groups as a consequence of different hematocrit targets at the start of CPB (Table 3) and that, additionally, resulted in significant differences in the volume of Gelofusine and FFP. In both groups, no crystalloids was used as the volume therapy during CPB. The mean ultrafiltration volume removed in the study group was 171 ± 99 ml (min 35 ml, max 480 ml, median 150 ml). No significant differences between the groups were found with regard to the total amount of allogeneic blood transfusion in the study period (study group 216 ± 92 ml versus control group 191 ± 93 ml; p = 0.136, Figure 1). During the ICU stay, only 2 out of 55 patients in the study group received allogeneic blood whereas 52 out of 70 patients in the control group were transfused with homologous RBC. Post-CPB transfusion of FFP, platelet concentrate and autologous blood product did not differ significantly between the groups; the study group received a total transfusion of 236 ± 204 ml versus 211 ± 198 ml (p = 0.26) in the control group. The groups did not differ significantly in respect to the preoperative values of Hb, Hct, Plt (Figure 2 a, b, c) and Lac (control group 1.4 ± 1.0 mmol/l versus study group 1.2 ± 0.5 mmol/l, p = 0.35). At the end of CPB, the study group had a significantly higher hematocrit than the control group and this significant difference in favour of the study group remained till the end of the operation. However, after 24 hours, the groups did not differ in respect to the hematocrit value. Plasma lactate concentration at the start of CPB (control group 1.6 ± 0.7 mmol/l versus study group 1.7 ± 0.8 mmol/l, p = 0.30) and at the end of operation (control group 1.6 ± 1.0 mmol/l versus study group 1.7 ± 1.1 mmol/l, p = 0.49) did not show significant differences between the groups. The COP in the study group did not change significantly during CPB (19.8 ± 2.1 mmHg at the start and 19.3 ± 2.3 mmHg at the end of bypass, p = 0.76). The COP of the control group diminished significantly during CPB from 19.4 ± 1.9 mmHg at the start to 17.6 ± 2.1 mmHg at the end of CPB, p = 0.003). Therefore, at the end of bypass, the COP of the study group was significantly higher compared to the control group (19.3 ± 2.3 mmHg versus 17.6 ± 2.1 mmHg, p
