Study Design And Participants
This single-institution cross-sectional study included 305 ambulatory female patients who had undergone surgery for breast cancer between 1988 and 2019. These patients attended the outpatient clinic at the Department of Breast Surgery at Nagoya City University Hospital in Japan. The eligibility criteria for inclusion in this study were: 1) patients who underwent breast cancer surgery at our institution, 2) patients who were disease-free survivors at least 1 year after breast cancer surgery, and 3) patients aged ≤ 65 years at the time of this study. We restricted the age of participating patients because younger age has been reported to be a predictor of CPSP[5] and the patient questionnaire was completed using an iPad (Apple Inc., Cupertino, CA, USA). The exclusion criteria were: 1) patients with severe pain due to diseases other than breast cancer, 2) patients who were treated for synchronous cancers or metachronous cancers, and 3) patients considered physically or mentally incapable of participation by their oncologists.
Procedure
Eligible patients were listed in advance and briefed about the study by their physician at the time of their outpatient visit. After informed consent had been obtained, patients were asked to complete the study questionnaire using an iPad at the hospital. The protocol for this study (60-19-0031) was approved by the Institutional Review Board of Nagoya City University Graduate School of Medical Sciences. This study conformed to the guidelines of the Declaration of Helsinki.
Data Collection
Demographic data and patient characteristics such as age (currently and at the time of surgery), time (months) since surgery, body mass index (BMI), type of surgery and reconstruction, stage of breast cancer at diagnosis, and type of adjuvant treatment were retrieved from patients’ medical records. A self-administered questionnaire was used to obtain additional information such as overall fear of breast cancer recurrence, pain severity and pain interference with life, distribution of postsurgical pain regions, use of analgesics after surgery, presence of psychiatric disorders, presence of family members, tobacco use, and alcohol intake. The presence of CPSP and need for treatment of CPSP were assessed with specific questions: 1) “Where is CPSP present?” (surgical site, axilla, arm, no pain) and 2) “Is there need for treatment of CPSP?” (yes, depends on the treatment, no). CPSP was defined as present when a patient selected one of the listed options (i.e., surgical site, axillary, arm). The need for CPSP treatment was defined as when a patient selected one of the listed options (i.e., yes, depends on treatment). The information obtained from the questionnaire was linked to the clinical information obtained from patients’ medical records.
Overall fear of breast cancer recurrence assessed by the Japanese version of the Concerns About Recurrence Scale (CARS-J)
The CARS is a 30-item breast cancer-specific self-report scale, originally developed in the USA[15]. It assesses the overall fear of breast cancer recurrence and covers four specific domains. Responses are on a scale from 0 to 6, with higher scores indicating greater overall fear of breast cancer recurrence. After obtaining permission to develop a Japanese version, a forward and back translation method was used to develop the CARS-J. The reliability and validity of the CARS-J has been confirmed among Japanese patients with breast cancer, although the factor structure differed slightly to that in the original study. This suggested there were some cross-cultural differences in the construct validity of fear of recurrence[16, 17]. In this study, we assessed overall fear of breast cancer recurrence using four items that covered frequency, potential for upset, consistency, and intensity of fear. The cut-off point for the CARS-J total score used to screen clinical psychological distress was defined as 11, as described elsewhere[18].
Brief Pain Inventory (bpi) And Pain Interference
The BPI has been shown to be a valid and reliable measure for assessing the intensity of postsurgical pain and pain interference in multiple aspects of life[19]. In this study, we used a short version of the BPI that uses a 24-hour recall period for patients developing CPSP. The BPI is a widely used measurement tool for assessing clinical pain and has proven reliability and validity[19–21]. We assessed the average pain severity in the last 24 hours, rated on a scale from 0 to 10. Higher scores indicate worse pain. The BPI pain interference (BPI-PI) subscale consists of seven items that assess enjoyment of life, general activity, walking ability, mood, sleep, normal work, and relationships with other people. Responses are on an 11-point scale from 0 = “no interference” to 10 = “complete interference”[22].
Statistical Analyses
The associations between patient characteristics and the presence of CPSP and the need for CPSP treatment were assessed using Student t-, χ2, and Fisher’s exact probability tests. The association between the presence of CPSP or need for CPSP treatment and overall fear of breast cancer recurrence scores (CARS-J) or pain severity and pain interference scores (BPI, BPI-PI) were assessed by Student t-tests. Because a cut-off value for the BPI-PI total score that predicts patients who need treatment for CPSP has not yet been reported, we performed receiver operating characteristic (ROC) curve analysis to determine the optimal cut-off point for the BPI-PI total score. We used Youden’s index (sensitivity + specificity – 1), which corresponded to a point on the ROC curve with the highest vertical distance from the 45° diagonal line.
Univariate and multivariate logistic regression models were developed to predict patients who develop CPSP and those who need treatment for CPSP. Variables included in the univariate logistic regression model were age at surgery, time (months) since surgery, BMI at surgery, use of analgesics within the first month after surgery, type of surgery, axillary lymph node dissection procedure, breast reconstruction, chemotherapy, radiotherapy, presence of family members, and presence of psychiatric disorders. Age at surgery and time since surgery were continuous variables. These factors were determined based on risk factors for CPSP reported in previous studies[5, 6, 11, 23–27]. We also included the distribution of postsurgical pain regions, CARS-J total score, and BPI-PI total score in the univariate analyses, as these factors may affect the presence of CPSP and the need for CPSP treatment. Variables for the multivariate logistic regression model were selected from the variables included in the univariate logistic regression model using a stepwise approach. The model predictive value was assessed with the C-statistic (area under the curve) in the ROC analysis. The level of statistical significance was set at a P-value less than 5%. All statistical calculations were performed with R software version 4.0.0 (https://www.R-project.org/).