This retrospective observational longitudinal study analyzed data sequentially extracted from a control spreadsheet created and archived by the Psychology Service of the Heart Institute, from March 2018 to March 2020. The psychological evaluations were conducted at three points: preoperative in the outpatient clinic, immediate postoperative (after discharge from the Intensive Care Unit and before hospital discharge), and late postoperative (6 months after hospital discharge).

The assessments were conducted within the Surgical Valve Disease Care Line (Figure 1), a structured clinical pathway implemented by the institution to standardize preoperative and postoperative follow-up.

The Care Line Model is a structured, multidisciplinary protocol that integrates sequential assessments by cardiology, surgery, nursing, rehabilitation teams, and psychology. Unlike traditional fragmented follow-up, this model ensures continuous, interdisciplinary monitoring of patients from preoperative evaluation to long-term postoperative recovery. It also standardizes the timing and content of psychological assessments, enabling consistent measurement of patient-reported outcomes over time.

The team involved includes clinical cardiologists, surgeons, anesthetists, dentists, nurses, pharmacists, nutritionists, social workers, physiotherapists, psychologists, and other professionals. Patients were eligible if they were indicated for valve intervention, regardless of waiting time, number of previous surgeries, or valve prosthesis type (biological or mechanical), were between 18 and 65 years old, had any level of education, and were followed by the Heart Institute outpatient clinic after intervention.

Exclusion criteria were:

Patients are not evaluated by psychology at all three points.

A 36-item self-report questionnaire assessing eight domains of health-related quality of life. Scores range from 0 to 100, with higher scores indicating better perceived health (19) (Supplementary Material).

Assesses five dimensions (mobility, self-care, usual activities, pain/discomfort, anxiety/depression) using three-level responses, plus a visual analogue scale (0–100) (20) (Supplementary Material).

In the descriptive analysis, data were expressed as absolute and relative frequencies for categorical variables and as mean and standard deviation for numerical variables. The probability distribution models adopted in the analyses were defined according to data adherence based on the Quasi-Likelihood under Independence Model Criterion (QIC). According to this criterion, a lower QIC value indicates better data adherence to the model (21).

The evolution of quality of life after valve surgical intervention was assessed using Generalized Estimating Equations (GEE) for the SF-36 scale domains. Since the SF-36 consists of discrete numerical variables, normal and Poisson distribution models were compared. The Poisson distribution showed the lowest QIC values in all models performed (Table 1).

For the EQ-5D scale and its dimensions, which are continuous variables, normal and Gamma distribution models were compared. Since the Gamma distribution only allows positive values greater than zero and the EQ-5D scale and its dimensions presented negative or zero values, the EQ-5D scale and its dimensions were initially transformed for model comparison (Table 2). This transformation involved adding a constant to all variable values, defined as one plus the minimum value of the variable. The normal distribution presented the lowest QIC values in all analyzed models (Table 2). As the normal distribution model demonstrated better adherence, the analyses were conducted with the real (non-transformed) variables.

The adjusted (multivariate) GEE model was applied to variables that were statistically significant (p < 0.05). Tukey's multiple comparisons were used when necessary. In all analyses, the significance level was set at 5%, and the analyses were conducted using R statistical computing software version 4.0.5 (R Foundation for Statistical Computing, Vienna, Austria) (22).

Patients reported improvements in several dimensions of quality of life after valve surgery, as evidenced by EQ-5D results (Table 4; Figures 2, 3). The total EQ-5D score increased from 0.587 preoperatively to 0.723 in the late postoperative period, indicating a global enhancement in perceived health status.

Improvements were observed in the following areas:

Mobility: scores decreased (indicating fewer mobility problems), from 0.045 preoperatively to 0.024 in late postoperative follow-up.

Table 5 and Figure 4 illustrate significant improvements across nearly all SF-36 domains in the late postoperative period.

Key findings include:

Physical functioning: increased from 38.0 preoperatively to 71.9 6 months post-discharge.

Instruments were based on their strong psychometric properties, their wide adoption in cardiovascular outcomes research, and their ability to capture both general and health-specific quality of life domains. Additionally, both instruments allow comparison with international datasets and are validated for use in Brazilian populations. Studies highlight that QoL is a multifaceted construct, and no single instrument fully captures its complexity, hence the relevance of using two complementary scales (29, 30).

The difficulty in defining and operationalizing QoL, highlighted in classical works such as Gill and Feinstein (30) and reaffirmed in more recent analyses (31), underscores the importance of using reliable methodological frameworks when assessing patient trajectories. By applying two validated instruments at three different postoperative stages, this study adds methodological rigor to the understanding of QoL in valvular disease (9, 30–32).

Sociodemographic characteristics, particularly education, occupation, and gender, play a significant role in shaping QoL outcomes. Evidence from congenital and acquired heart disease indicates that socioeconomic status affects patients' ability to adapt to surgical recovery, access rehabilitation resources, and sustain long-term lifestyle changes. Our sample profile, which included predominantly women, individuals with low educational level, and a significant proportion of homemakers, reflects a sociodemographic configuration that may influence perceptions of recovery and the psychosocial demands of postoperative adjustment. These dimensions warrant further study and should be considered when designing individualized follow-up strategies within care pathways (33).

The present findings are consistent with systematic reviews and clinical studies reporting that valve surgery leads to substantial improvements in functioning, vitality, and symptom burden. Postoperative gains in mobility, self-care, and emotional health observed here parallel gains demonstrated in large observational courts and randomized studies (6, 34–37).

More recent investigations (2021–2025) emphasize early mobilization, functional recovery trajectories, and patient-reported outcomes following cardiac surgery. These studies reinforce the role of structured and interdisciplinary care models in optimizing postoperative QoL. Incorporating these perspectives strengthens the relevance of our findings and situates the present study within contemporary international discussions (38–40).

The Care Line Model represents an innovative contribution by institutionalizing interdisciplinary follow-up and embedding psychological assessment into routine perioperative care. The structured sequence of appointments ensures that QoL data is collected consistently at clinically meaningful moments, enabling the detection of psychosocial needs that may emerge during recovery. This framework not only improves data quality but also reinforces patient-centered care, aligning with value-based health care principles (41, 42).

Given the global emphasis on patient-centered outcomes and integrated care pathways, our findings support the integration of structured models into cardiovascular surgical programs, particularly in settings where fragmentation of care persists (43, 44).

All analyses were interpreted as discrete comparisons between the three time periods. The improvements observed at 6 months should not be understood as linear or progressive trends throughout the postoperative period, but rather as cumulative effects of surgery, rehabilitation, and psychosocial adaptation. This approach reinforces the clinical importance of longitudinal rather than continuous assessment models to ensure long-term recovery. Psychosocial support is particularly relevant given the fluctuations observed in anxiety and depression scores, emphasizing that emotional adaptation may not follow the same trajectory as physical recovery. Incorporating psychological assessment into routine postoperative care may facilitate the early identification of vulnerabilities and improve overall quality-of-life outcomes.

This study's limitations include its single-center design and restricted sample size, although the repeated-measures approach enhances internal validity.

Future research should expand on the interaction between socioeconomic variables and postoperative QoL, examine differential effects across types of valve interventions, and compare structured care pathways with standard follow-up models. Additionally, qualitative approaches may enrich understanding of the subjective meaning of recovery from the patient's perspective.