The present retrospective cohort study was conducted in Shiraz Breast Clinic, Shiraz, Iran. The mentioned center is the main referral clinic for patients with breast cancers in the South of Iran. In the current study, the clinical data of patients with the impression of Invasive ductal carcinoma who underwent breast cancer surgery were assessed between 2006, and 2016. All studied patients were followed up till 2020. Patients were excluded from the study if met the following criteria: bilateral breast cancer, immediate breast reconstruction surgery, metastatic breast cancer, other malignancy, history of breast surgery, administration of both IV and inhalation anesthetics, male gender, benign breast tumor or carcinoma in situ, American Society of Anesthesiologists [13] physical status greater than or equal to IV, and unknown type of anesthesia or receiving local and regional anesthesia. Patients were categorized based on the received anesthesia into the Total IV (IV group which was treated with continuous infusions of propofol and remifentanil) or volatile/inhalation anesthesia (the volatile/inhalational group which was treated with sevoflurane or isoflurane) and additional analgesic opioids if necessary. The administered anesthesia was selected by the attending anesthesiologists based on the patient’s condition.

The following data were obtained from the recorded data in the breast cancer registry: surgery type, recurrence condition, pre- or post-operation adjuvant chemotherapy, administered radiotherapy or endocrine therapy, invasion condition, tumor size, estrogen and progesterone receptor status, histological tumor grade, histopathological type of tumor, epidermal growth factor receptor type 2 (HER-2) expression, and level of axillary node involvement.

Data regarding the type and the dose of administered analgesics and opioids were gathered through reviewing the hospital records of patients who underwent surgery in Shahid Faghihi hospital affiliated with Shiraz University of medical science. The duration of the operation, duration of anesthesia, and date of surgery were also noted. In this study for reducing the effect of the confounding factors, all the patients were matched according to age, tumor grade, tumor size and hormonal receptor.

The current study results are described as the number and percentage for categorical variables and as the mean ± SD for continuous variables. At first, the normality of the variables was checked through the Kolmogorov–Smirnov test. Considering that the significance value was greater than 0.05. Therefore, all variables have a normal distribution. The independent samples T-test was utilized to assess continuous variables. Also, the Chi-square test was applied in a bid to investigate categorical variables between groups.

In this cohort study, recurrence-free survival and overall survival were calculated for up to 2 and 5 years by conducting the Kaplan–Meier method, and the groups were compared by the log-rank test. The full Cox proportional hazards models were utilized for univariate and multivariate analysis of variables influencing recurrence of breast cancer; potential risk factors such as age, duration of anesthesia, type of anesthesia, involvement of right or left breast, tumor size, tumor grade, in situ component, tumor necrosis, type of lymph node involvement, margin involvement, axillary management, ER, PR, Her2, neoadjuvant and adjuvant chemotherapy, radiotherapy, IORT, and hormone therapy. In addition, backwards multivariable selection were performed. In the backward selection method, all predictive variables are first entered into the equation, and then if they do not meet the criteria to remain in the model, they are removed one by one from the model. Furthermore, the considerable variables (P < 0.5) of univariate analysis were considered in multivariate analysis. At the end of this statistical analysis, propensity score matching was applied to decrease the possible confounding impacts of each variable and the variances in baseline characteristics between the groups. The variables used for matching were age, tumor size, and tumor grade. The applied matching method were propensity score matching. It should be noted that these variables were selected based on the previous literature. In this regard, the result of match analysis was like that of the initial analysis. All statistical analyses were conducted with SPSS software version 25.