Diagnostic evaluation of risk for bleeding in cardiac surgery with extracorporeal circulation*

ABSTRACT Objective: to identify the risk factors associated with cases of excessive bleeding in patients submitted to cardiac surgery with extracorporeal circulation. Method: case-control study on the factors of risk for bleeding based on the analysis of data from the medical charts of 216 patients submitted to cardiac surgery with elective extracorporeal circulation during a three-year period. Results: variables that are commonly associated with excessive bleeding in studies in the field were analyzed, and the following were considered as risk factors for the nursing diagnosis “risk for bleeding” (00206) in cardiac surgery with extracorporeal circulation: Body mass index lower than 26.35kg/m² (Odds ratio = 3.64); Extracorporeal circulation longer than 90 minutes (Odds ratio = 3.57); Hypothermia lower than 32°C (Odds ratio = 2.86); Metabolic acidosis (Odds ratio = 3.50) and Activated partial thromboplastin time longer than 40 seconds (Odds ratio= 2.55). Conclusion: such variables may be clinical indicators of an operational nature for a better characterization of the risk factor “treatment regimen” and a refinement of knowledge related to coagulopathy induced by extracorporeal circulation, which is currently presumably incorporated into the “treatment regimen” category of the nursing diagnostic classification by NANDA International, Inc.


Method
This is a case-control study on the factors of risk for bleeding conducted at a general federal tertiary military hospital located in the city of Rio de Janeiro, southeastern Variables potentially associated with postoperative bleeding were selected from the taxonomy of NANDA International, Inc. (8) and from a literature review.
The preoperative variables selected for the study as risk factors are presented in Figure 1.
The intraoperative variables selected for the study as risk factors are presented in Figure 2.
The postoperative variables selected for the study as risk factors are presented in Figure 3.
Studies and clinical practice have adopted different definitions and criteria to define abnormal perioperative bleeding, such as drainage through the thoracic tubes, magnitude of blood transfusions, delayed sternal closure, and use of coagulation concentrates (2) . The present study adopted the concept of excessive bleeding as a function of the volume drained by the thoracic tubes because it is aligned with a nursing intervention and more easily articulated to the nursing diagnosis. The following were adopted as criteria for excessive bleeding: bleeding from mediastinal and/or pleural tubes with values equal to or greater than 1.5ml drainage/kg/h for at least 3 hours (11) , or drainage greater than 200ml/h or fraction of an hour or greater than 2ml/kg/hour for two consecutive hours in the first 6 hours postoperatively (3) .

Variable
Operational definitions and criteria of risk for excessive bleeding ECC time* Obtained from the perfusion form filled out by the perfusion professional in the intraoperative period. Risk for bleeding: prolonged ECC*, of approximately 90-110 minutes (16)(17)(18) .

Surgery time
Obtained from the intraoperative control form filled out by the nursing technician circulating in the room, which shows the time of surgery. Risk for bleeding: time longer than the average time of 300 minutes for cardiac surgery (19) .
Minimum esophageal temperature (reduced) Collected from the perfusion form filled out by the perfusion professional. Esophageal temperature reached during ECC* time. Risk for bleeding: lower than 32°C † (20)(21) .

Heparin reinforcement in ECC*
Obtained from the perfusion form filled out by the perfusion professional. Risk for bleeding: an additional dose of heparin (5000IU ‡ ), from 1ml § , administered by the perfusion professional into the ECC* circuit when evaluating the ACT || test.  were not available for consultation were discarded. There were three death losses during the perioperative period; therefore, 230 charts remained.
Based on the excessive-bleeding criteria, the occurrence of 24 cases was observed, and the case/ control ratio was established as 1 case to 8 controls, so as to maximize the statistical power of the study in view of the available sample (22)(23) . One hundred and ninetytwo medical charts were drawn as control.
The chi-square tests or Fisher's test were used to test the differences between the proportions obtained for the groups of cases and controls. In order to evaluate the risk of the variables for bleeding, the odds ratios (OR) and the 95% confidence intervals were calculated.
The variables that obtained an odds ratio above 1.0 with p < 0.05 were considered as risk factors with * Medical Records Departaments statistical significance. The data were analyzed by the MedCalceasy-to-use statistical software®.

Discussion
The study found variables that were associated with excessive bleeding after cardiac surgery with ECC in the preoperative and intraoperative periods and in the first postoperative minutes. BMI lower than 26.35kg/ m 2 , ECC time longer than 90 minutes, esophageal temperature lower than 32°C, and metabolic acidosis and activated partial thromboplastin time longer than 40s were validated factors and that have already been identified in other studies (2)(3)(4)6,9,12,24) . On the other hand, there were variables that did not reach statistical significance values in the present study, but were considered to be associated with excessive bleeding in the abovementioned investigations. closure and evaluation of the use of transfusions (2)(3)9) .
The results obtained in the present study were encouraging in relation to the criterion adopted to define excessive bleeding, especially because it was observed that all the patients who needed surgical reexploration were in the case-group. In a way, this shows a practical value of the criterion selected to establish the phenomenon of excessive bleeding.
Regarding BMI values, the studies present different cut-off points to establish the value that defines the excessive-bleeding condition in the postoperative period or with hemorrhagic complications, such as for example: lower than 20kg/m 2 (24) , lower than or equal to 24kg/m 2 (15) , 25±3kg/m 2(12) and lower than 26.35kg/ m 2 (9) . We chose to use the value lower than 26.35 kg/ m 2 as a characterization of the risk factor when considering the means for the case-and control-groups.
Once the variable in question was tested, an increased odds ratio was obtained with statistical significance, which was essential to consider it a diagnostic risk factor. Despite the differences among the BMI values that would be related to excessive bleeding, there is something in common among the studies, namely: the predictive character that low weight increases the risk for bleeding. Using the results of the abovementioned studies, an assumption is made that patients submitted to ECC are more susceptible to the effects of changing coagulation factors during hemodilution (6) . Consideration should be given to the possible risks of associating low BMI with high crystalloid infusion, as an example, of the implications for the professionals involved, including perfusion professionals.
The relationships in the increased postoperative bleeding attributed to risk factors of low esophageal temperature, metabolic acidosis and changes in activated partial thromboplastin time are undesired consequences of extracorporeal circulation (6) . ECC produces a set of responses related to the interaction involving inflammatory reactions, fibrinolysis and coagulation, and hemodilution and increased consumption of coagulation factors due to increased fibrinolysis may be the cause of ECC-induced coagulopathy (25) .
Metabolic acidosis and hypothermia induced by ECC contribute to exacerbate changes in the coagulation chain (25) . Such alterations associated with the other risk factors probably close a multifactor mechanism that culminates in the increased mean of activated partial thromboplastin time (PTTa), which was observed in the postoperative period of patients in the case-group.
PTTa is one of the laboratory tests that comprise the evaluation standard for management of post-ECC coagulopathy (26) and, in light of the results, it becomes a clinical risk-factor indicator to be evaluated for the definition of the nursing diagnosis risk for bleeding in the postoperative period.
The heparin reinforcement administered, on average, at a larger dose to the case-group when compared to the control-group (Table 1) may be related to excessive bleeding. This hypothesis gains strength when we consider that a longer ECC time also increases the need for heparin administration (heparin reinforcement) (6) .
It is understood that the study brings relevant contributions to the refinement of nursing diagnosis risk for bleeding, as it allows for the application of the concept in the field of cardiac surgery by providing operational elements for the best use of risk factor "treatment regimen" and clinical adequacy of ECCinduced coagulopathy as a substitute for risk factor "inherent coagulopathy" (8) . Coagulopathy is due to coagulation disorders that include complications related to trauma or are inherent to the patient, such as thrombocytopenia, for example (8,27) , and it cannot be means of clinical studies (validation and testing), level 3 of evidence (8) .
For the nursing diagnostic evaluation, the study supports the relevance of measurement or monitoring actions that can be performed by the nursing team, It is understood that the main limitation of the study was related to its retrospective nature, especially regarding the potential bias of information inaccuracy.
However, the low occurrence of bleeding at the institution chosen for the study was one of the criteria that motivated the choice of the case-control research design. The authors assume that the data production conditions, already presented in the method section, may have minimized such methodological limit, which is difficult to overcome in the type of study developed.

Conclusion
Considering the findings in the present study, it was concluded that the variables associated with excessive bleeding after cardiac surgery with extracorporeal circulation were: BMI lower than 26.35kg/m 2 , ECC time longer than 90 minutes, esophageal temperature lower than 32°C and metabolic acidosis and activated partial thromboplastin time longer than 40s. Such variables can be considered as clinical indicators that would best characterize risk factor "treatment regimen" of the diagnostic classification by NANDA International, Inc.
for the clientele studied. In addition, it would support the delimitation of elements for operational definitions related to ECC-induced coagulopathy, considering that the nursing diagnosis risk factor "inherent coagulopathy" does not adequately apply to most cases of excessive bleeding in the postoperative period of cardiac surgery.
The study on the factors of risk for bleeding in cardiac surgery provides relevant information for the validation of the nursing diagnosis and application in the care for people undergoing cardiac surgeries.