Background: Intraoperative and postoperative morbi-mortality factors are multiple in pediatric patients. Studies in pediatric cardiac surgery and intensive care patients have identified transfusion as one independent factor among others. There is not a lot of data concerning transfusion related morbi-mortality in other pediatric patients fields like neurosurgery, abdominal and orthopedic surgery. These were investigated.

Objectives: To identify morbi-mortality risk factors in intraoperatively transfused and not transfused pediatric patients in neurosurgery, abdominal and orthopedic surgery.

Design: Retrospective observational descriptive pediatric cohort study.
Setting: Monocentric pediatric tertiary center, Necker Enfants Malades University Hospital Paris, from 1 January 2014 to 17 Mai 2017.

Patients: 594 patients with mean age of 90.86±71.80 months were included. Inclusion criteria were the presence or the absence of transfusion in the intraoperative period in neurosurgery, abdominal and orthopedic surgery patients. Exclusion criterion was transfusion in the postoperative period until discharge from hospital.

Main outcome measures: Primary outcome was mortality and secondary outcome was morbidity in transfused and non-transfused patients. Mortality was assessed by deaths occuring intraoperatively or postoperatively during the entire hospitalisation. Morbidity was assessed by intraoperative, postoperative complications, repeat surgery, length of stay in the intensive care unit, in the hospitalisation ward, total length of stay in hospital and length of mechanical ventilation.

Results: Multivariate analysis revealed that ASA score was the independent risk factor for mortality. Transfusion, emergency surgery, type of surgery, age and prematurity were independent risk factors for morbidity.

Conclusion: Patient outcome can be improved by applying specific preventive measures on each risk factor.

In pediatric patients admitted for surgery under anesthesia, morbi-mortality is related to mutiple factors. Several morbi-mortality risk factors have been identified of which transfusion is one of the independent risk factors in studies concerning pediatric cardiac surgery and critical care patients.^1-3 This study was undertaken to determine whether transfusion is an independant morbi-mortality risk factor in three different pediatric surgical populations: neurosurgery, orthopedics and abdominal surgery. The primary endpoint was to identify factors related to mortality and the secondary endpoint was to identify factors related to morbidity in this pediatric population. Mortality (primary outcome) was assessed by deaths occurring intraoperatively or postoperatively until discharge from hospital. Morbidity (secondary outcome) was assessed by intraoperative and postoperative complications, repeat surgery, length of stay in the intensive care unit (LOSICU), length of stay in hospital (LOSHOSP), total length of stay in hospital (intensive care and standard hospitalisation ward, TLOSHOSP) and length of mechanical ventilation (LMV).

After approval from the Ethics Committee of Necker Enfants Malades University Hospital, Paris, France, under the registration number 2017-CK-5-R1 on 21 March 2017 (Chairperson Professor Mariane de Montalembert) and after declaration of this study to the National Commission of Liberties and Computer Science, Paris, France (CNIL, Commission Nationale des Libertés et de l’Informatique) under the registration number 2028257 v0 on 21 February 2017 (Chairperson Mrs Isabelle Falque Pierrotin), 594 patients with mean age of 90.86±71.80 (± standard deviation) months where included in this study from our Hospital, Necker Enfants Malades, Paris. Inclusion criteria consisted of patients admitted for neurosurgery, orthopedic and abdominal surgery and who received blood products [packed red blood cells (PRBC) and/or fresh frozen plasma (FFP) and /or concentrated platelet units (CPU)] in the intraoperative period (transfusion group,) and patients admitted for the same surgical specialities and who did not receive any blood transfusion during surgery or in the postoperative period.

We first included the transfused patients and then patients who did not receive blood components, in order to include patients with same surgical operations whenever possible. The local Transfusion Department (EFS, Establishment Français de Sang, HôpitalUniversitaire Necker Enfants Malades) provided a list of patients who had been transfused in the operation theater from 1 January 2014 until 31 December 2016. There were 1500 transfused patients identified of which only 292 were finaly retained for the study because of complete data and also in order to have the same number of patients with equivalent surgical operations as in the no transfusion group. We used the operation theater programmation system (IPOP) to identifiy patients who did not receive blood products intraoperatively and postoperatively. We included 302 patients from 1 January 2014 until 17 Mai 2017 in the no transfusion group. Whenever possible, patients scheduled for similar interventions as in the transfusedgroup where included.

Medical records were analyzed using the computer medical report system (Orbis, Mediweb and Cristalnet). Data collected consisted of intraoperative and postoperative mortality occuring during hospitalisation regardless of TLOSHOSP (to assess primary outcome), intraoperative and postoperative complications which included organe failure and infections, repeat surgery, number of days spent in the intensive care unit and in the hospitalisation ward, total number of days spent in hospital, number of days spent under mechanical ventilation (to assess secondary outcome). Factors that could influence primary and secondary outcomes were collected: age, prematurity, type of surgery, comorbidities, ASA score (American Society of Anesthesiologists Score), emergency surgery, number of units of blood products administered [packed red blood cell units (PRBC), fresh frozen plasma units (FFP), concentrated platelet units (CPU], preoperative and postoperative hemoglobin and platelet concentration. The ASA score (I-V) is a scale used in anesthesia to assess patient severity physical status: ASA I: normal healty patient, ASA II: patient with mild systemic disease, ASA III: patient with severe systemic disease, ASA IV: patient with severe systemic disease which is constantly threatening life, ASA V: moribund patient who is not expected to survive without surgery. Missing data concerning patient weight, intraoperative blood loss and fluid therapy with cristalloids and colloids, coagulation analysis like international normalized ratio, activated partial thromboplastine time, fibrinogen blood levels which could influence blood transfusion were not taken into account since they were not always available.

XLSTAT 2017.4 software was used for statistics. Statistical tests included Student’s test for parametrical variables, Chi square or Fischer’s exact test to compare category variables; propensity score matching analysis to assess for cofounding morbi-mortality factors, logistic, log-linear regressions and multiple factor analysis (MFA) for multivariate analysis. We considered significant a p-value equals to or less than 0.05. We firstly identified risk factors with univariate analysis in each subgroup (neurosurgery, orthopedic and abdominal surgery) and then we included the three groups subgroups in one single group. Secondly we proceeded with multivariate measures: logistic, log-linear regressions and to confirm the results obtained by these, MFA was finaly realized.⁴ Mean values were expressed with standard deviation (±SD) and category values are expressed as proportions.

Six risk factors were identified (ASA score, emergency surgery transfusion (units of blood products administered PRBC+FFP+CUP), age, prematurity and type of surgery) and correlated to the number deaths (mortality) during hospitalisation, number of patients with intraoperative and postoperative complications (complications), repeat surgery, number of days spent in the intensive care unit (LOSICU), in the hospitalisation ward (LOSHOSP), total number of days spent in hospital (ICU plus hospitalisation ward, TLOSHOSP), and the number of days spent under mechanical ventilation (LMV). Hemoglobin and platelet concentration were not taken into account for analysis since some of the data was not available.

We included 594 patients, 292 transfused patients and 302 patients without transfusion. Table 1 illustrates some characteristics in the transfused and non-transfused patients in univariate analysis. The number of patients with complications, of repeat surgery, deaths, emergency surgery, LOSHOSP, TLOSHOSP, LMV were higher in the transfused group (T). The proportion of patients with higher ASA scores (³ IV) was bigger in the T group. Non premature patients were younger in the T group but the premature patients were concentrated in the non transfused group (NT). Table 2 illustrates, different surgical interventions. 60 different types of interventions were identified. The most frequent surgical intervention in the T group was scoliosis 63/292) followed by craniosynostosis (48/292), liver transplantation (19/292), intracerebral tumor exeresis (16/292), intestinal exeresis (14/292), neuroblastoma (13/292) and limb tumor exeresis (11/292). In the NT group, the most frequent surgery was scoliosis (61/302) followed by intracerebral tumor exeresis (38/302), intestinal exeresis (23/302), limb tumor exeresis (22/302), craniosynostosis (21/302), kidney transplantation (11/302), esophageal atresia (14/302) pelvic tumor (10/302),Chiari’s Malformation (10/302).

Table 3 illustrates the type of blood components in the transfusion group. There were 780 units of blood components (PRBC+ FFP+ CPU) administered amoung 292 patients. The most administered blood products were PRBC units (481/780), followed by FFP units (226/780 and CPU (73/780). Table 4 illustrates patients’ comorbidities. 79 comorbidities were identified. In the T group, the most frequent comorbidities were hepatic failure (25/292), cerebral anoxic lesions (19/292), intracerebral tumor (14/292), Ewing sarcoma (14/292) and cancer (12/292). In the NT group, the most frequent comorbidities were intracerebral tumor (36/302), cancer (15/302), prematurity (13/302), epilepsia (13/302), Chiari’s malformation type I (12/302) and hemorrhagic diathesis (10/302). Table 5 illustrates intraoperative and postoperative complications. 25 different complications were identified. In the T group there were 220 complications and 53 in NT group. In the T group the most frequent intraoperative complication was hemorrhagic shock (22/220); the most frequent postoperative organe failure complications were neurologic (20/220), cardiocirculatory (17/220) and respiratory (13/220); the most frequent postoperative infectious complications were pulmonary (33/220), abdominal (23/220), urinary (12/220); there were 34 (34/220) repeat surgeries and 9 (9/220) deaths. In the NT group the most frequent intraoperative complication was cardiac arrest (2/53), the most frequent postoperative organe failures were respiratory (8/53), neurologic (5/53); the most frequent postoperative infectious complications were pulmonary (5/53) and surgical wound sepsis (4/53); there were 11 repeat surgeries (11/53) and 2 deaths (2/53).

Table 3 Blood component units in transfusion group

After multivariate analysis with logistic regression for complications, repeat surgery and mortality, results are shown in tables 6 to 8 and ROC curves are illustrated in Figures 1-3. Figures 4-6 illustrate the contributions of each risk factor to mortality, complications and to repeat surgery. ASA score is the independent risk factor for mortality in this cohort study (Table 6). ASA score, transfusion, emergency surgery and abdominal surgery are independent risk factors for complications (Table 7). Emergency surgery, transfusion and orthopedic surgery are independent risk factors for repeat surgery (Table 8). The ROC curves show that the area under the curve (AUC) equals 1 for mortality (Figure 1), 0.8 for complications and repeat surgery (Figures 2 & 3). The contribution of each risk factor is illustrated in Figure 4 (mortality), Figure 5 (complications) and Figure 6 (repeat surgery).

Figure 1 Mortality ROC curve.

Figure 2 Complications ROC curve.

Figure 3 Repeat surgery ROC curve.

Figure 4 Contribution of each risk factor to mortality.

Figure 5 Contribution of each risk factor to complications.

Figure 6 Contribution of each risk factor to complications.

Log linear regression was realised for LOSICU, LOSHOSP, TLOSHOSP and LMV (length of mechanical ventilation) and the results are shown in tables 9-12. ASA score, emergency surgery, transfusion, age, prematurity and abdominal surgery are independent risk factors for length of stay in the intensive care unit (Table 9). ASA score, transfusion, age and abdominal surgery are independent risk factors for LOSHOSP (Table 10). ASA score, emergency surgery, transfusion, age and abdominal surgery are independent risk factors for TLOSHOSP (Table 11). ASA score, emergency surgery, transfusion, age and prematurity are independent risk factors for length of mechanical ventilation (Table 12). Propensity score matching (PMS) analysis revealed that 100% of transfused patients 97% of non-transfused patients (Table 13), 100% of patients who died, 100% of those who presented with repeat surgery and 99% of patients with complications were matched in this cohort study. Multiple factor analysis (MFA) (Table 14) confirmed the results obtained with logistic and log linear regressions.

Our study has shown that in this pediatric retrospective cohort study, perioperative and postoperative mortality and morbidity were determined by multiple factors. We focused on some of these factors: transfusion, ASA score, emergency surgery, type of surgery, age and prematurity. Studies in pediatric cardiac surgery and critically ill pediatric patients have reported the role of transfusion as an independent morbi-mortality factor. This survey concerned critically ill pediatric patients from the intraoperative and the postoperative period to discharge from hospital. Transfusion was not found in our study to be an independent risk factor for mortality but as an independent risk factor for morbidity as assessed by intra and postoperative complications, repeat surgery, LOSICU, LOSHSOP, TLOSHOSP and LMV. ASA score (³III) was the independent risk factor for mortality. Univariate analysis showed that in the transfusion group there were more complications, more repeat surgery; mortality, LOSICU, LOSHOSP, TLOSHOSP LMV and ASA score were higher than in the no transfusion group. This could be explained in part by the selection bias of our study but still the same results were obtained when analyzing the three subgroups separately (neurosurgery, orthopedics and abdominal surgery) (data was not shown here to simplify the results) and propensity score matching analysis showed that 100% of the transfused and 97% non transfused patients were matched.

Nevertheless, patients who needed transfusion had more comorbidities, higher ASA scores thus were more critically ill and were much more exposed to hemorrhagic surgery (liver transplantation, craniosynostosis, scoliosis, trauma…). Multivariate analysis when considering all patients in the same group revealed other multiple independent morbidity factors: emergency surgery, type of surgery, age, and prematurity. Our study highlightened some factors which were only the visible part of the iceberg since perioperative and postoperative morbi-mortality is multifactorial. Other factors that could influence patient outcome like hemoglobin concentration, weight, fluid therapy with colloids and cristalloids, blood loss, type of anesthesia (all patients had general anesthesia in this study) and organisational aspects were not analysed here. LOSICU, LOSHOSP, TLOSHOSP and LMV depend upon several factors and those analysed in our study are far from being exhaustive but they can help to understand some aspects implicated in morbidity.

Patient outcome can be improved by reducing morbi-mortality risk factors. Since ASA score is the independent mortality risk factor in this study, perioperative management of high ASA score patients should be adapted to patient status and surgery. Optimizing transfusion strategies could improve patient outcome, studies have reported the physiopathology underlying some transfusion related complications.^5,6 Exposure to blood products can be reduced by applying restrictive transfusion strategies,⁷ using transfusion protocols based on bedside viscoelastic methods to guide blood components administration during hemorrhagic surgery like liver transplantation, craniosynostosis, scoliosis and trauma.^8-13 Emergency surgery is an independent risk factor for morbidity, only urgent interventions should be realised during emergency periods and non urgent operations realised electively.

Age is also an independent factor for morbidity. A recent prospective multicenter study reported the importance for management of pediatric patients under a certain age in specialiazed centers, the importance of a good training and supervision environment.¹⁴ Our survey enlarged the analysis on transfusion and morbi- mortality factors to critically ill pediatric patients in neurosurgery, abdominal and orthopedic surgery. Our survey enlarged the analysis on transfusion and morbi-mortality factors to critically ill pediatric patients in neurosurgery, abdominal and orthopedic surgery. We focused on these three groups since we did not find a lot of data concerning this subject in these particular fields.

We deliberately included the three subspecialities in one group because univariate analysis of each subgroup revealed the same results as the all in one group. We do not pretend to have analysed all morbi-mortality risk factors in our study but only some of those factors which were accessible and thus analysable. Identifying morbi-mortality factor is one of the first steps towards patient outcome improvement. Once these factors are identified, preventive measures can be applied. Our study had limits: it was retrospective, monocenter, some data concerning factors which could influence outcome was missing. More prospective multicenter studies are needed to analyse the invisible part of the iceberg which contains the multiple morbi-mortality factors.

Assistance with the article: We would like to thank Professor Philippe Van Der Linden and Professor Christian Mélot for review and advise of this article.

None.

1. Willemse A, Van Lerberghe C, Gonsette K, et al. The indication for perioperative red  blood cell transfusion is a predictive risk factor for severe postoperative morbidity in  children undergoing cardiac surgery. Eur J Cardiothorac Surg. 2014;45(6):1050‒1057.
2. Kneyber MCJ, Hersi MI, Twisk JWR, et al. Red blood cell transfusion in critically ill children is independently associated with increased mortlity. Intensive Care Med. 2007;33(8):1414‒1422.
3. Rajasekeran S, Kort E, Hacbarth R, et al. Red cell Transfusion as an independent risk for mortality in crtically ill children. J Intensive Care. 2016;4:2.
4. Mélot C. Les analyses multivariables. Rev Mal Respir. 2005;22(4):687‒690.
5. El Kenz H, Van der Linden P. Transfusion-related acute lung injury.  Eur J Anaesthesiol. 2013;30:1‒6.
6. Mulder HD, Augustijn QJ, Van Woensel JB, et al. Incidence, risk factors and outcome oftransfusion-related acute lung injury in critically ill children. J Crit Care. 2015;30(1):55‒59.
7. Lacroix J, Hébert PC, Hutchison JS, et al. Transfusion Strategies for Patients in Pediatric Intensive Care Units.  N Engl J Med. 2007;356:1609‒1619.
8. Kozek-Langenecker SA, Afshari A, Albaladejo P, et al. Management of severe  perioperative bleeding. Guidelines from the European Society of Anaesthesiology. Eur J Anaesthesiology. 2013;30(6):270‒382.
9. Kloesel B, Kovatsis PG, Faraoni D, et al. Incidence and predictors of massive bleeding in children undergoing liver transplantation : a single-center retrospective analysis. Paediatr Anaesth. 2017;27(7):718‒772.
10. Haas T, Goobie S, Spielmann N, et al. Improvements in patient blood management for  pediatric craniosynostosis surgery using ROTEM-assisted strategy-feasability and costs. Pediatr Anaesth. 2014;24(7):774‒780.
11. Bonfield CM, Sharma J, Cochrane DD, et al. Minimizing blood transfusions in the  surgical correction of craniosynostosis: a 10-year single–center experience. Childs Nerv Syst. 2016;32(1):143‒151.
12. Hassan N, Halanski M, Wincek J, et al. Blood management in pediatric spinal deformity surgery: review of a 2-year experience. Transfusion. 2011;51(10):2133‒2141.
13. Nystrup KB, Stensballe J, Bøttger M, et al. Transfusion therapy in paediatric trauma patients: a review of the literature. Scand J Trauma Resusc Emerg Med. 2015;23:21.
14. Habre W, Disma N, Virag K, et al. Incidence of severe critical events in paediatric anesthesia (APRICOT) : a prospective multicentre observational study in 261 hospitals in Europe.  Lancet Respir Med. 2017;5(5):412‒425.