Association of preoperative anxiety and depression symptoms with postoperative complications of cardiac surgeries

Objective to investigate the associations of preoperative anxiety and depression symptoms with postoperative complications and with sociodemographic and clinical characteristics of patients submitted to the first coronary artery bypass graft. Method observational, analytical and longitudinal study. A consecutive non-probabilistic sample consisted of patients submitted to coronary artery bypass graft. To evaluate the symptoms, the Hospital Anxiety and Depression Scale was used. tracheal intubation for more than 48 hours, hemodynamic instability, sensorineural deficit, agitation, hyperglycemia, infection, nausea, vomiting, pain and death were classified as complications. The Mann-Whitney and Spearman Correlation tests were used, with a significance level of 0.05. Results a total of 75 patients participated. The group that presented hemodynamic instability in the postoperative period had a greater median for the anxiety symptoms (p = 0.012), as well as the women (p = 0.028). The median of the depression symptoms was higher in the group presenting nausea (p = 0.002), agitation (p <0.001), tracheal intubation for more than 48 hours (p = 0.018) and sensorineural deficit (p = 0.016). Conclusion there was association of the symptoms of preoperative anxiety with hemodynamic instability in the postoperative period and with the female gender, as well as association of depression symptoms with the following complications: nausea, agitation, time of intubation in the postoperative period and sensorineural deficit.

influencing PO outcomes, increased risk of unwanted outcomes, as well as the length of hospital stay (10) .
In international studies, researchers investigated the association between preoperative anxiety and depression symptoms with PO complications, with a focus on mortality up to 30 days after surgery (11)(12)(13)(14) . In the present study, we chose to investigate the presence of complications and death in the PO period, during the patients' stay in the Intensive Care Unit (ICU). This is justified by the fact that this is the period of greatest instability and criticality, as well as due to the scarce scientific production focusing on this period. Patients who did not present cognitive conditions to answer the questionnaires or who presented clinical decompensation of heart disease on the day of data collection in the preoperative period (presence of dyspnea, precordialgia or orotracheal intubation) were excluded.
To identify the patients with cognitive conditions to answer the questionnaires, we used the instrument "Mini Mental State Examination -MMSE" (15) , in the adapted version to the Portuguese language (16) . The cut-off points adopted in this study were as follows: illiterate patients had to score at least 13 points; those with one to seven years of schooling, a minimum of 18 points; and with eight or more years of schooling, at least 26 points (17) .
In the preoperative period, the data were collected for the sociodemographic and clinical characterization by means of an interview with the patients and consultation of the medical charts, besides the evaluation of anxiety and depression symptoms through an interview conducted on the day before the surgery.
After a maximum of 24 hours after discharge from the ICU, the data related to the evolution of the patient in the PO period were collected by consulting the medical charts and the electronic data system used in the ICU Rodrigues HF, Furuya RK, Dantas RAS, Rodrigues AJ, Dessotte CAM.
of the said hospital. The PO period investigated in this study refers to the period of stay of the patients in the ICU, that is, from the moment of admission, after the end of the surgery, until the discharge of the unit.
For the characterization of PO complications, we analyzed the patient's evolutions performed by the nursing and medical teams and the nursing care instrument in the PO period, which are routinely filled in the referred institution. The results of serum lactate levels and central venous oxygen saturation were obtained from the electronic data system.
For the characterization of the participants, an instrument was elaborated based on the literature review and in an earlier study (9) , containing sociodemographic data (birth, hospitalization and interview dates, sex, marital status, schooling, professional status and monthly family income) and clinical data (presence of associated pathologies, left ventricular ejection fraction, smoking, use of psychotropic drugs before the surgery and rescheduling of surgery). The age was calculated by subtracting the date of the interview from the date of birth, since the preoperative admission time was calculated by subtracting the date of the surgery from the admission date.
For the evaluation of left ventricular ejection fraction (LVEF), we considered values greater than or equal to 50% for preserved LVEF, and less than 50% for decreased LVEF (18) . These values are also used as reference in the hospital where the study was developed.
The following data were collected from the PO should present serum lactate values below <2 mmol/L and central venous oxygen saturation above >70% (21) .
The patient was classified as having "hemodynamic instability" when a concomitant alteration of three of the described parameters was observed at any time during the period of their stay in the ICU, regardless of the number of times this condition was observed.
The Hospital Anxiety and Depression Scale (HADS) instrument was used to evaluate the symptoms of preoperative anxiety and depression (22) , in its version adapted to Portuguese language (23) . The HADS values between 0.30-0.50 indicate moderate correlation and above 0.50, strong correlation (24) . The level of significance was set at 0.05.  Table 2.  The mean length of stay in the ICU was 3.5 days (SD = 2.3, median = 3.0), ranging from 2 to 18 days.
Regarding the preoperative symptoms of anxiety and depression, the studied patients presented a mean of 5.5 (SD = 4.6, median of 4.0) for the anxiety symptoms, ranging from zero to 17. For the depression symptoms, the mean values were 4.0 (SD = 3.9, median 3.0), ranging from zero to 18.       Table 5. The occurrence of death in the postoperative period, during the patients' stay in the ICU, among patients submitted to CABG, was 1.3%. As the patient was not followed throughout the whole PO period and patients who died in the operating room were not included, the results of this study cannot be compared with mortality rates found in the international and national literature.
In any case, this is a very relevant topic because, in clinical practice, patients' emotional symptoms are poorly evaluated, especially when considering the severity of the cardiovascular disease and the presence of numerous patient comorbidities in the perioperative period of these surgeries.
An international study was carried out in Australia with 158 patients undergoing CABG, with or without concomitant valve procedures, in which the authors showed that generalized anxiety in the preoperative period was a significant predictor for the occurrence of cardiovascular and cerebrovascular complications in the PO period (14) .
In a study conducted in Germany, the authors evaluated the efficacy of a short-term, informationbased intervention and emotional support to reduce preoperative anxiety and in the PO period of patients undergoing CABG. They found that the group of patients who received the intervention reported a moderately lower state of anxiety after the intervention and prior to surgery, as well as in the PO period, five days after surgery, with statistically significant differences between groups (p <0.001); however, it did not show a significant difference in mortality in the PO period and in the length of stay in the ICU (11) .
In Sweden, researchers retrospectively evaluated 56,064 patients who had undergone CABG between 1997 and 2008. The objective was to determine the possible association of preoperative depression symptoms with long-term survival, that is, until the occurrence of death from any cause or readmission due to myocardial infarction, heart failure or stroke. They found that 324 (0.6%) patients had preoperative depression. During an average follow-up of 7.5 years, 35% of these patients with depression evolved to death, compared to 25% in the non-depressed group. They concluded that preoperative depression was significantly associated with increased long-term mortality (13) .
Researchers from a major American study concluded that depression is an important predictor of mortality following cardiac surgeries, regardless of risk factors, such as smoking, age, sex, diabetes mellitus, prior myocardial infarction, number of grafts and LVEF, since, after adjustment for these factors, depression was associated with a two-to three-fold increase in mortality risk (25) . Depression was also pointed out in other studies as an independent risk factor for mortality after CABG (26)(27) .
The present study also proposed to evaluate the association of preoperative anxiety and depression symptoms with the sociodemographic characteristics (sex, age, marital status and professional status) of the patients. It was found that women had the highest median for anxiety symptoms when compared with men, and the difference found was statistically significant, that is, women had more preoperative anxiety symptoms than men.
The reasons that may explain the increased frequency of anxiety symptoms among women are  (12)(13)(31)(32) , being the female sex a risk factor for the development of depression before the CABG (31) . These differences may result from the application of several instruments for the evaluation of the symptoms, as well as the heterogeneity of the moments in which the data were collected, that is, with which anticipation from the surgery the symptoms were evaluated.
In this study, no differences were found in the evaluation of anxiety and depression symptoms in relation to age, marital status and employment status.
However, a higher frequency of anxiety and depression symptoms was previously documented among younger patients undergoing CABG (27,32) . It has also been identified that patients undergoing CABG who are supported by their families, especially their spouses, recover more quickly than those who do not have such support (33) , thus evidencing the impact of the marital status on the coping of the patient submitted to the surgery. In the same way, it is already described in the literature that the preoccupation with work or with being away from it is a factor that bothers the patients in the preoperative period of CABG (34) .
Regarding the correlations between preoperative anxiety and depression symptoms and preoperative admission time, surgery time and length of stay in the ICU in the subjects submitted to CABG, the correlations found in this study were weak and without statistical significance. In a study carried out in Greece, the authors also did not find a relation between the symptoms of preoperative depression and the preoperative admission time, corroborating with the results of this research (32) .
On the other hand, in a study developed in Spain, the authors evidenced that the preoperative admission time greater than three days was a determining factor for the onset of depression symptoms, before the CABG (27) .