No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Predicting Reduction Mammaplasty Total Resection Weight With Machine Learning
Abstract
Background
Machine learning (ML) is a form of artificial intelligence that has been used to create better predictive models in medicine. Using ML algorithms, we sought to create a predictive model for breast resection weight based on anthropometric measurements.
Methods
We analyzed 237 patients (474 individual breasts) who underwent reduction mammoplasty at our institution. Anthropometric variables included body surface area (BSA), body mass index, sternal notch–to–nipple (SN-N), and nipple–to–inframammary fold values. Four different ML algorithms (linear regression, ridge regression, support vector regression, and random forest regression) either including or excluding the Schnur Scale prediction for the same data were trained and tested on their ability to recognize the relationship between the anthropometric variables and total resection weights. Resection weight prediction accuracy for each model and the Schnur scale alone were evaluated based on using mean absolute error (MAE).
Results
In our cohort, mean age was 40.36 years. Most patients (71.61%) were African American. Mean BSA was 2.0 m ² , mean body mass index was 33.045 kg/m ² , mean SN-N was 35.0 cm, and mean nipple–to–inframammary fold was 16.0 cm. Mean SN-N was found to have the greatest variable importance. All 4 models made resection weight predictions with MAE lower than that of the Schnur Scale alone in both the training and testing datasets. Overall, the random forest regression model without Schnur scale weight had the lowest MAE at 186.20.
Conclusion
Our ML resection weight prediction model represents an accurate and promising alternative to the Schnur Scale in the setting of reduction mammaplasty consultations.
This systematic review aims to assess the use of machine learning, deep learning, and artificial intelligence in aesthetic plastic surgery.
This qualitative systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta‐analyses reporting guideline. To analyze quality risk-of-bias assessment of all included articles, we used the ROBINS-I tool for non-randomized studies. We searched for studies with the following MeSH terms: Machine Learning OR Deep Learning OR Artificial intelligence AND Plastic surgery on MEDLINE/PubMed, EMBASE, and Cochrane Library, from inception until July 2024 without any filter applied.
A total of 2,148 studies were screened and 41 were fully reviewed. We conducted article extraction, screening, and full text review using the rayyan tool. Eighteen studies were ultimately included in this review, describing the use of machine learning, deep learning and artificial intelligence in aesthetic plastic surgery. All studies were published from 2019 to 2024. Articles varied regarding the population studied, type of machine learning (ML), Deep Learning Model (DLM), Artificial Intelligence (AI) used, and aesthetic plastic surgery type. Of the eighteen studies, we included the following aesthetic plastic surgeries: augmentation mastopexy, breast augmentation, reduction mammoplasty, rhinoplasty, facial rejuvenation surgery, including facelift surgery; blepharoplasty, and body contouring. Image-based with AI, ML, and DLMs algorithms were used in these studies to improve human decision-making and identified factors associated with postoperative complications.
AI, ML, and DL algorithms offer immense potential to transform the aesthetic plastic surgery field. By meticulously analyzing patient data, these technologies may, in the future, help optimize treatment plans, predict potential complications, and more clearly elucidate patient concerns, improving their ability to make informed decisions. The drawback, as with preoperative surgical simulation, is that patients may see an AI-generated image that is to their liking, but impossible to achieve; great care is needed when using such tools in order to not create unrealistic expectations.
Ultimately, the old plastic surgery adage of ‘’under-promise and over-deliver’’ will continue to hold true, at least for the foreseeable future.
This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266.
Study registration A review protocol for this systematic review was registered at PROSPERO CRD42024567461.
Background:
It is important to be able to predict, for each individual patient, the likelihood of later metastatic occurrence, because the prediction can guide treatment plans tailored to a specific patient to prevent metastasis and to help avoid under-treatment or over-treatment. Deep neural network (DNN) learning, commonly referred to as deep learning, has become popular due to its success in image detection and prediction, but questions such as whether deep learning outperforms other machine learning methods when using non-image clinical data remain unanswered. Grid search has been introduced to deep learning hyperparameter tuning for the purpose of improving its prediction performance, but the effect of grid search on other machine learning methods are under-studied. In this research, we take the empirical approach to study the performance of deep learning and other machine learning methods when using non-image clinical data to predict the occurrence of breast cancer metastasis (BCM) 5, 10, or 15 years after the initial treatment. We developed prediction models using the deep feedforward neural network (DFNN) methods, as well as models using nine other machine learning methods, including naïve Bayes (NB), logistic regression (LR), support vector machine (SVM), LASSO, decision tree (DT), k-nearest neighbor (KNN), random forest (RF), AdaBoost (ADB), and XGBoost (XGB). We used grid search to tune hyperparameters for all methods. We then compared our feedforward deep learning models to the models trained using the nine other machine learning methods.
Results:
Based on the mean test AUC (Area under the ROC Curve) results, DFNN ranks 6th, 4th, and 3rd when predicting 5-year, 10-year, and 15-year BCM, respectively, out of 10 methods. The top performing methods in predicting 5-year BCM are XGB (1st), RF (2nd), and KNN (3rd). For predicting 10-year BCM, the top performers are XGB (1st), RF (2nd), and NB (3rd). Finally, for 15-year BCM, the top performers are SVM (1st), LR and LASSO (tied for 2nd), and DFNN (3rd). The ensemble methods RF and XGB outperform other methods when data are less balanced, while SVM, LR, LASSO, and DFNN outperform other methods when data are more balanced. Our statistical testing results show that at a significance level of 0.05, DFNN overall performs comparably to other machine learning methods when predicting 5-year, 10-year, and 15-year BCM.
Conclusions:
Our results show that deep learning with grid search overall performs at least as well as other machine learning methods when using non-image clinical data. It is interesting to note that some of the other machine learning methods, such as XGB, RF, and SVM, are very strong competitors of DFNN when incorporating grid search. It is also worth noting that the computation time required to do grid search with DFNN is much more than that required to do grid search with the other nine machine learning methods.
Objective:
This study aimed to investigate whether there was an increase in the number of postoperative complications in patients undergoing reduction mammoplasty depending on the technique used (i.e., pedicle type, skin incision pattern), existence of concomitant diseases, and presence of other risk factors.
Materials and methods:
A total of 186 patients who underwent breast reduction between 2013 and 2018 (bilateral, n = 170; unilateral, n = 16) were included in the study. A retrospective review of the data of patients who underwent reduction mammoplasty, which was performed by the same surgical team in a single institution over a 6-year period, was carried out. Superomedial, superior, and inferior pedicles were used in 99, 55, and 32 patients, respectively. The median follow-up period was 4 years.
Results:
The median patient age was 45 (range: 16-75) years. The median total reduction weight was 2,194 (range: 80-4,800) grams. The median distance between the sternal notch and nipple was 31 cm (range: 24-45 cm) for the right breast and 30 cm (range: 22-45 cm) for the left breast. The overall complication rate was 6.9%. The complication rates in patients with and without any concomitant diseases were 10.2% and 4.6%, respectively. The overall complication rate was significantly higher in patients with smoking habit, accessory breasts, progesterone use, cerebrovascular disease, morbid obesity (Body Mass Index ≥40 kg/m2), and thalassemia.
Conclusion:
Our analysis shows that the presence of concomitant diseases increases the risk for postoperative complications in patients who underwent reduction mammoplasty. Our findings do not suggest that any of the techniques have significant superiority to each other in terms of pedicle safety and overall complication rate.
Clinical prediction models aim to predict outcomes in individuals, to inform diagnosis or prognosis in healthcare. Hundreds of prediction models are published in the medical literature each year, yet many are developed using a dataset that is too small for the total number of participants or outcome events. This leads to inaccurate predictions and consequently incorrect healthcare decisions for some individuals. In this article, the authors provide guidance on how to calculate the sample size required to develop a clinical prediction model.
Background:
Supervised machine learning algorithms have been a dominant method in the data mining field. Disease prediction using health data has recently shown a potential application area for these methods. This study ai7ms to identify the key trends among different types of supervised machine learning algorithms, and their performance and usage for disease risk prediction.
Methods:
In this study, extensive research efforts were made to identify those studies that applied more than one supervised machine learning algorithm on single disease prediction. Two databases (i.e., Scopus and PubMed) were searched for different types of search items. Thus, we selected 48 articles in total for the comparison among variants supervised machine learning algorithms for disease prediction.
Results:
We found that the Support Vector Machine (SVM) algorithm is applied most frequently (in 29 studies) followed by the Naïve Bayes algorithm (in 23 studies). However, the Random Forest (RF) algorithm showed superior accuracy comparatively. Of the 17 studies where it was applied, RF showed the highest accuracy in 9 of them, i.e., 53%. This was followed by SVM which topped in 41% of the studies it was considered.
Conclusion:
This study provides a wide overview of the relative performance of different variants of supervised machine learning algorithms for disease prediction. This important information of relative performance can be used to aid researchers in the selection of an appropriate supervised machine learning algorithm for their studies.
Background
Many insurance carriers continue to deny coverage for reduction mammaplasty unless a minimum amount of resected breast tissue per breast is achieved during surgery. This study investigates the accuracy of preoperative prediction that a minimum weight of 500 g can be resected and evaluates potential risk factors for not meeting this insurance requirement.
Methods
A retrospective review was performed on 445 patients with bilateral symptomatic macromastia who sought consultation for breast reduction surgery from 2007 to 2012. Women were included for analysis if they had documented predicted resection weights and underwent small-to-moderate breast reduction (< 1,000 g per side; n = 323). Relevant demographic information, mean predicted resection weight, and the mean actual resection weight were collected for analysis.
Results
Surgeon prediction of resection weight being over 500 g had a positive predictive value of 73%. In 61 patients (19%), the predicted weights were ≥ 500 g, but the actual weights were < 500 g. Thirty percentage of these 61 patients did not meet either Schnur or minimum weight requirements. Women with a body mass index < 30 were at significantly increased odds (odds ratio, 3.76; 95% confidence interval, 1.89–7.48; P = 0.002) of not meeting the minimum weight requirement at surgery compared with patients with a body mass index ≥ 30.
Conclusions
The common insurance criterion of removing ≥ 500 g per breast during breast reduction surgery are not met in a distinct cohort of women who are clinically appropriate candidates. This risk is particularly increased in nonobese women possibly due to proportionately smaller breast mass compared with obese women.
Importance:
Atrial fibrillation (AF) contributes to substantial morbidity, mortality, and health care expenditures. Accurate prediction of incident AF would enhance AF management and potentially improve patient outcomes.
Objective:
To validate the AF risk prediction model originally developed by the Cohorts for Heart and Aging Research in Genomic Epidemiology-Atrial Fibrillation (CHARGE-AF) investigators using a large repository of electronic medical records (EMRs).
Design, setting, and participants:
In this prediction model study, deidentified EMRs of 33 494 individuals 40 years or older who were white or African American and had no history of AF were reviewed and analyzed. The participants were followed up in the internal medicine outpatient clinics at Vanderbilt University Medical Center for incident AF from December 31, 2005, until December 31, 2010. Adjusting for differences in baseline hazard, the CHARGE-AF Cox proportional hazards model regression coefficients were applied to the EMR cohort. A simple version of the model with no echocardiographic variables was also evaluated. Data were analyzed from October 31, 2013, to January 31, 2014.
Main outcomes and measures:
Incident AF. Predictors in the model included age, race, height, weight, systolic and diastolic blood pressure, treatment for hypertension, smoking status, type 2 diabetes, heart failure, history of myocardial infarction, left ventricular hypertrophy, and PR interval.
Results:
Among the 33 494 participants, the median age was 57 (interquartile range, 49-67) years; 57% of patients were women, 43% were men, 85.7% were white, and 14.3% were African American. During the mean (SD) follow-up of 4.8 (0.9) years, 2455 individuals (7.3%) developed AF. Both models had poor calibration in the EMR cohort, with underprediction of AF among low-risk individuals and overprediction of AF among high-risk individuals (10th and 90th percentiles for predicted probability of incident AF, 0.005 and 0.179, respectively). The full CHARGE-AF model had a C index of 0.708 (95% CI, 0.699-0.718) in our cohort. The simple model had similar discrimination (C index, 0.709; 95% CI, 0.699-0.718; P = .70 for difference between models).
Conclusions and relevance:
Despite reasonable discrimination, the CHARGE-AF models showed poor calibration in this EMR cohort. This study highlights the difficulties of applying a risk model derived from prospective cohort studies to an EMR cohort and suggests that these AF risk prediction models be used with caution in the EMR setting. Future risk models may need to be developed and validated within EMR cohorts.
Risk prediction plays an important role in clinical cardiology research. Traditionally, most risk models have been based on regression models. While useful and robust, these statistical methods are limited to using a small number of predictors which operate in the same way on everyone, and uniformly throughout their range. The purpose of this review is to illustrate the use of machine-learning methods for development of risk prediction models. Typically presented as black box approaches, most machine-learning methods are aimed at solving particular challenges that arise in data analysis that are not well addressed by typical regression approaches. To illustrate these challenges, as well as how different methods can address them, we consider trying to predicting mortality after diagnosis of acute myocardial infarction. We use data derived from our institution's electronic health record and abstract data on 13 regularly measured laboratory markers. We walk through different challenges that arise in modelling these data and then introduce different machine-learning approaches. Finally, we discuss general issues in the application of machine-learning methods including tuning parameters, loss functions, variable importance, and missing data. Overall, this review serves as an introduction for those working on risk modelling to approach the diffuse field of machine learning.
Medical decision-making is increasingly based on quantifiable data. From the moment patients come into contact with the health care system, their entire medical history is recorded electronically. Whether a patient is in the operating room or on the hospital ward, technological advancement has facilitated the expedient and reliable measurement of clinically relevant health metrics, all in an effort to guide care and ensure the best possible clinical outcomes. However, as the volume and complexity of biomedical data grow, it becomes challenging to effectively process "big data" using conventional techniques. Physicians and scientists must be prepared to look beyond classic methods of data processing to extract clinically relevant information. The purpose of this article is to introduce the modern plastic surgeon to machine learning and computational interpretation of large data sets. What is machine learning? Machine learning, a subfield of artificial intelligence, can address clinically relevant problems in several domains of plastic surgery, including burn surgery; microsurgery; and craniofacial, peripheral nerve, and aesthetic surgery. This article provides a brief introduction to current research and suggests future projects that will allow plastic surgeons to explore this new frontier of surgical science.
When the number of events is low relative to the number of predictors, standard regression could produce overfitted risk models that make inaccurate predictions. Use of penalised regression may improve the accuracy of risk prediction
Scikit-learn is an increasingly popular machine learning li- brary. Written
in Python, it is designed to be simple and efficient, accessible to
non-experts, and reusable in various contexts. In this paper, we present and
discuss our design choices for the application programming interface (API) of
the project. In particular, we describe the simple and elegant interface shared
by all learning and processing units in the library and then discuss its
advantages in terms of composition and reusability. The paper also comments on
implementation details specific to the Python ecosystem and analyzes obstacles
faced by users and developers of the library.
Reduction mammaplasty is a frequently performed plastic surgical procedure, yet preoperative assessment usually excludes an estimated resection weight. In this retrospective study, a formula was derived using regression analysis of routinely measured preoperative anthropomorphic measurements and intraoperative resection weights.
Data were collected from 214 consecutive breast reduction patients operated on at a tertiary referral center between January of 1993 and January of 2006. The operations were performed by 15 plastic surgeons using a variety of reduction pattern techniques. Four variables were recorded for each patient: notch to nipple distance, nipple to inframammary crease distance, body mass index, and age. Regression analysis was performed on the data to establish which preoperative measurements correlated most accurately with the resection weight.
The following formula was established relating nipple to inframammary crease distance and notch to nipple distance measurement to the weight removed: Breast weight = (35.4 x notch to nipple distance + 60.66 x nipple to inframammary crease distance) - 1239.64.
Use of this formula aids the surgeon with patient counseling, insurance company quotes, and intraoperatively in patients with asymmetry or those undergoing reconstruction, as a guide to resection weights.
Background
Racial and socioeconomic disparities in access and quality of surgical care are well documented in many surgical subspecialties, including plastic surgery.
Objectives
Here, we aim to determine if demographic disparities exist in preoperative and postoperative satisfaction after breast reduction mammaplasty, using patient-reported quality of life (QoL) scores.
Methods
Patients who underwent breast reduction mammaplasty between 2015 – 2020 were identified. Patients that underwent complex concomitant procedures were excluded. Patient demographics and QoL, as measured by the BREAST-Q, were extracted. Wilcoxon Rank Sum and Kruskal Wallis tests were used to compare QoL scores across demographic subgroups.
Results
115 patients met inclusion criteria. QoL improved across all four BREAST-Q domains (all p<0.001). Disparities were shown to exist in the following: median income vs. postoperative satisfaction with information (p<0.001), BMI vs. preoperative physical well-being (p<0.001), and ethnicity vs. preoperative physical well-being (p=0.003). A sub-group analysis of Caucasian patients compared to Black/African American patients revealed significant inequalities in BMI (p<0.001), median income by zip code (p<0.001), improvement in satisfaction with breasts (p=0.039), satisfaction with information (p=0.007), and satisfaction with office staff (p=0.044).
Conclusions
Racial and socioeconomic inequalities exist in terms of preoperative and postoperative satisfaction for patients undergoing breast reduction mammaplasty. Institutions ought to focus on developing tools for equitable and inclusive patient education and perioperative counseling.
Background
Reduction mammoplasty is a common procedure associated with a very high patient satisfaction rate. It has been shown to alleviate symptoms related to macromastia, such as back, neck, and shoulder pain, poor posture, bra strap grooving, paresthesia, and rashes. Despite the manifold benefits of reduction mammaplasty, some insurance companies require minimum resection weights of at least 500-grams per breast in order to distinguish between a reconstructive and aesthetic procedure.
Objective
The aim of this study is to assess the origins of the 500-gram rule used in reduction mammaplasty.
Methods
A comprehensive literature search of the MEDLINE, PubMed, Google Scholar, EMBASE, and Cochrane Central Register of Controlled Trials was conducted for studies published through July 2020 with multiple search terms related to resection weight criteria for breast reduction. Data on criteria, outcomes, and patient satisfaction were collected.
Results
A total of 14 articles were selected from the 27 articles that were identified. 500-gram rule appears to be arbitrary. It appears that it is not based on any available hard evidence. However, numerous studies show that patients who have less than 500-grams of tissue removed from each breast still experience significant symptomatic relief from reduction mammaplasty with a marked improvement in quality of life.
Conclusion
The 500-gram rule should be re-evaluated as criterion for pre-authorization reduction mammaplasty for insurance companies. It may influence surgeons to choose between form and function. Many additional patients might then benefit from this procedure with significant benefits.
Background:
Reduction mammaplasty provides symptomatic relief to patients with macromastia. Insurance companies act as gatekeepers of health care by determining the medical necessity of surgical procedures, including reduction mammaplasty. The authors sought to evaluate insurance coverage and policy criteria for reduction mammaplasty.
Methods:
The authors conducted a cross-sectional analysis of U.S. insurance policies on reduction mammaplasty. Insurance providers were selected based on their enrolment data and market share. The authors conducted telephone interviews and Web-based searches to identify the policies. Medical necessity criteria were abstracted from the publicly available policies that offered coverage.
Results:
The authors reviewed 63 insurers. One in 10 insurers had no established policy for reduction mammaplasty. Of the 48 publicly available policies, shoulder pain and backache were the most common symptoms required for preapproval (98 percent and 98 percent). A minimum resection volume was requested by 88 percent of policies. One-third of policies (31 percent) offered a choice between removal of a minimum weight per breast or a volume based on body surface area. Over half of companies (54 percent) used body surface area calculations to predict minimum resection volume. Medical necessity that extended beyond national recommendations included trial of weight loss (23 percent) and nipple position (10 percent).
Conclusions:
Insurance policy criteria for reduction mammaplasty are discordant with current national recommendations and current clinical evidence. Many policies use outdated criteria that do not correlate with symptom relief and consequently limit access to reduction mammaplasty. Here, the authors propose a comprehensive guideline to maximize coverage of reduction mammaplasty.
Background:
Reduction mammaplasty is the standard of care for symptomatic macromastia. The process of requesting insurance coverage for reduction mammaplasty is cumbersome and potentially controversial, and insurance policies vary significantly. The goal of our analysis is to identify trends in insurance coverage rates, assess for the presence of disparities, and propose ways to increase chances of successful preauthorization.
Methods:
The authors performed a retrospective review of preauthorizations for reduction mammaplasty at a single institution from 2012 to 2017. Insurance company names were deidentified. Preauthorization denial rates were assessed by year, insurance carrier, and reason for denial. Multiple regression analysis was performed to identify predictors for predetermination denial by insurance companies.
Results:
Among 295 preauthorizations, 212 were approved (72 percent) and 83 were denied (28 percent), among which 18 were appealed, 13 successfully. Rates of insurance denials have been increasing steadily, from 18 percent to 41 percent. Medicaid had the lowest denial rates (9.3 percent), whereas private carriers denials ranged from 21.4 to 62.1 percent. In terms of reason for denial, 30 percent were because of contract exclusion, 39 percent were because of inadequate documentation or not meeting medical criteria, and 12 percent were because of inadequate predicted resection weight. Certain private insurance carriers were the only independent predictors of predetermination denial.
Conclusions:
Rate of preauthorization denials is high and has been increasing steadily. Insurance criteria remain arbitrary. A proper documentation and appeal process by the plastic surgeon may improve rates of insurance approval. Although resection weight does not correlate with symptom relief, predicted breast tissue resection weight continues to be critical for insurance approval.
This Perspective explores the application of machine learning toward improved diagnosis and treatment. We outline a vision for how machine learning can transform three broad areas of biomedicine: clinical diagnostics, precision treatments, and health monitoring, where the goal is to maintain health through a range of diseases and the normal aging process. For each area, early instances of successful machine learning applications are discussed, as well as opportunities and challenges for machine learning. When these challenges are met, machine learning promises a future of rigorous, outcomes-based medicine with detection, diagnosis, and treatment strategies that are continuously adapted to individual and environmental differences.
While radiologists regularly issue follow-up recommendations, our preliminary research has shown that anywhere from 35 to 50% of patients who receive follow-up recommendations for findings of possible cancer on abdominopelvic imaging do not return for follow-up. As such, they remain at risk for adverse outcomes related to missed or delayed cancer diagnosis. In this study, we develop an algorithm to automatically detect free text radiology reports that have a follow-up recommendation using natural language processing (NLP) techniques and machine learning models. The data set used in this study consists of 6000 free text reports from the author’s institution. NLP techniques are used to engineer 1500 features, which include the most informative unigrams, bigrams, and trigrams in the training corpus after performing tokenization and Porter stemming. On this data set, we train naive Bayes, decision tree, and maximum entropy models. The decision tree model, with an F1 score of 0.458 and accuracy of 0.862, outperforms both the naive Bayes (F1 score of 0.381) and maximum entropy (F1 score of 0.387) models. The models were analyzed to determine predictive features, with term frequency of n-grams such as “renal neoplasm” and “evalu with enhanc” being most predictive of a follow-up recommendation. Key to maximizing performance was feature engineering that extracts predictive information and appropriate selection of machine learning algorithms based on the feature set.
Patient satisfaction and outcomes following reduction mammoplasty is important to measure, being a being a reconstructive procedure with physical and cosmetic benefits. This study aimed to evaluate patient satisfaction and the various questionnaires that have been devised for this measurement. A systematic search of literature was performed in PubMed, Cochrane Library, Medline and Scopus databases from 1966 to July 2018 according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. After application of pre-determined inclusion criteria by two authors, 95 articles were included. Data was extracted from included studies relating to demographics, surgical technique, questionnaires used and physical, psychological and aesthetic outcomes. Of the 95 studies included (9,716 patients), 58 studies (5,867 patients) reported on overall satisfaction with a mean rate of 90.26%. Researchers' own non-validated questionnaire was most commonly used in 52.6% of studies. Validated questionnaires used were most commonly the SF-36 (25.3%), Rosenberg self-esteem scale (RSES) (9.5%) and BREAST-Q (8.4%). All showed improvement in physical and mental health. Our findings suggest that women who have undergone reduction mammoplasty for breast hypertrophy report postoperative satisfaction and improvement in quality of life. Of the validated questionnaires used, a combination of those assessing both mental, physical and psychosocial health as well as breast-specific surveys were most commonly used and may provide an accurate assessment of patient outcomes.
Background:
Reduction mammaplasty is a highly effective procedure for treatment of symptomatic macromastia. Prediction of resection weight is important for the surgeon and the patient, but none of the current prediction models is widely accepted. Insurance carriers are arbitrarily using resection weight to determine medical necessity, despite published literature supporting that resection weight does not correlate with symptomatic relief. What is the most accurate method of predicting resection weight and what is its role in breast reduction surgery?
Methods:
The authors conducted a retrospective review of patients who underwent reduction mammaplasty at a single institution from 2012 to 2017. A senior biostatistician performed multiple regression analysis to identify predictors of resection weight, and linear regression models were created to compare each of the established prediction scales to actual resected weight. Patient outcomes were evaluated.
Results:
Three-hundred fourteen patients were included. A new prediction model was created. The Galveston scale performed the best (R = 0.73; p < 0.001), whereas the Schnur scale performed the worst (R = 0.43; p < 0.001). The Appel and Descamps scales had variable performance in different subcategories of body mass index and menopausal status (p < 0.01). Internal validation confirmed the Galveston scale's best predictive value; 38.6 percent and 28.9 percent of actual breast resection weights were below Schnur prediction and 500-g minimum, respectively, yet 97 percent of patients reported symptomatic improvement or relief.
Conclusions:
The authors recommend a patient-specific and surgeon-specific approach for prediction of resection weight in breast reduction. The Galveston scale fits the best for older patients with higher body mass indices and breasts requiring large resections. Medical necessity decisions should be based on patient symptoms, physical examination, and the physician's clinical judgment.
Clinical question/level of evidence:
Diagnostic, IV.
Introduction:
Breast hypertrophy is a condition associated with physical, psychological, and psychosocial problems. The primary aims of this study were to determine the impact of breast hypertrophy and the effects of breast reduction, performed on the basis of well-described inclusion criteria, on general and breast-related health, using both general and diagnosis-specific validated questionnaires. We used a prospective, longitudinal paired study design. A secondary aim was to analyze the relationship between preoperative breast volume, body mass index, sternal notch-to-nipple distance and the weight of resected tissue on the one hand and improvements in health on the other.
Methods:
Three hundred forty-eight consecutive patients undergoing breast reduction were included and the Short-Form 36 (SF-36), Breast-Related Symptoms Questionnaire (BRSQ), Modified Breast Evaluation Questionnaire (mBEQ) and BREAST-Q were distributed preoperatively and 1 year postoperatively.
Results:
A total of 284 (83%) patients answered the questionnaires either preoperatively or postoperatively, or both, and 159 (46%) patients answered both. The breast hypertrophy patients had significantly lower scores preoperatively than the matched normal population when it came to all dimensions of the SF-36 and mBEQ. The preoperative scores for both the BRSQ and BREAST-Q were low.After breast reduction, there were significant improvements in all dimensions of the BRSQ, mBEQ, and Breast-Q and in several dimensions of the SF-36.
Conclusions:
Breast reduction reduces or removes disease-associated pain. It improves or normalizes perceived health and psychosocial self-esteem in slightly obese women or women of normal weight with preoperative breast volumes around 1000 mL. Women with higher preoperative breast volumes and longer sternal notch-to-nipple distances appear to be more satisfied with the cosmetic result postoperatively.
Recent years have seen tremendous progress in artificial intelligence (AI), such as with the automatic and real-time recognition of objects and activities in videos in the field of computer vision. Due to its increasing digitalization, the operating room (OR) promises to directly benefit from this progress in the form of new assistance tools that can enhance the abilities and performance of surgical teams. Key for such tools is the recognition of the surgical workflow, because efficient assistance by an AI system requires this system to be aware of the surgical context, namely of all activities taking place inside the operating room. We present here how several recent techniques relying on machine and deep learning can be used to analyze the activities taking place during surgery, using videos captured from either endoscopic or ceiling-mounted cameras. We also present two potential clinical applications that we are developing at the University of Strasbourg with our clinical partners.
Background:
Reduction mammaplasty is known for excellent outcomes and patient satisfaction. Although patients report improvements in pain, weight loss, and exercise levels, objective data on physical fitness benefits are limited.
Methods:
Using the Stanford Military Data Repository, we identified 89 US Army active duty women with at least 1 pre- and postoperative Army Physical Fitness Test (APFT) who underwent reduction mammaplasty during 2011 to 2014. We used paired t tests to compare pre- and postoperative APFT score means and raw values for push-ups, sit-ups, and the 2-mile run.
Results:
There were 56 subjects (62.9%) who improved in total APFT scores. Total score means increased from 235.9 preoperatively to 243.4 postoperatively (P = 0.0065). Of 28 subjects with at least 2 APFT scores before and after surgery, 20 (71.4%) improved in total scores. The subgroup's mean total score increased from 237.8 to 251.3 (P = 0.0009). Comparing individual pre- and postprocedure APFTs, all subjects demonstrated a mean 3.9% (SD, 0.1) improvement in total scores, and the subpopulation of 28 improved by 6.3% (SD, 0.1). In all events, mean performance values trended toward better postoperative scores. Differences were statistically significant for the total population for the number of sit-ups (P = 0.035), and, for the subgroup of 28, differences were statistically significant for the total score (P = 0.0009), sit-ups (P = 0.0002), and push-ups (P = 0.0134).
Conclusions:
Reduction mammaplasty was associated with postoperative physical fitness improvements among US Army active duty women. Soldier data are useful for objectively assessing physical fitness effects of breast reduction surgery.
Learning Objectives: After reading this article, the participant should be able to: 1. Understand the multiple reduction mammaplasty techniques available for patients and describe the advantages and disadvantages associated with each. 2. Describe the indications for the treatment of macromastia in patients younger than 18 years. 3. Identify the preoperative indications for breast imaging before surgery. 4. Describe the benefits of breast infiltration with local anesthesia with epinephrine before surgery. 5. Understand the use of deep venous thrombosis prophylaxis in breast reduction surgery. 6. Describe when the use of drains is indicated after breast reduction surgery.
Summary: The goal of this Continuing Medical Education module is to summarize key evidence-based data available to plastic surgeons to improve their care of patients with breast hypertrophy. The authors’ goal is to present the current controversies regarding their treatment and provide a discussion of the various options in their care. The article was prepared to accompany practice-based assessment with ongoing surgical education for the Maintenance of Certification Program of the American Board of Plastic Surgery.
Background Machine-learning models may aid cardiac phenotypic recognition by using features of cardiac tissue deformation.
Objectives This study investigated the diagnostic value of a machine-learning framework that incorporates speckle-tracking echocardiographic data for automated discrimination of hypertrophic cardiomyopathy (HCM) from physiological hypertrophy seen in athletes (ATH).
Methods Expert-annotated speckle-tracking echocardiographic datasets obtained from 77 ATH and 62 HCM patients were used for developing an automated system. An ensemble machine-learning model with 3 different machine-learning algorithms (support vector machines, random forests, and artificial neural networks) was developed and a majority voting method was used for conclusive predictions with further K-fold cross-validation.
Results Feature selection using an information gain (IG) algorithm revealed that volume was the best predictor for differentiating between HCM ands. ATH (IG = 0.24) followed by mid-left ventricular segmental (IG = 0.134) and average longitudinal strain (IG = 0.131). The ensemble machine-learning model showed increased sensitivity and specificity compared with early-to-late diastolic transmitral velocity ratio (p < 0.01), average early diastolic tissue velocity (e′) (p < 0.01), and strain (p = 0.04). Because ATH were younger, adjusted analysis was undertaken in younger HCM patients and compared with ATH with left ventricular wall thickness >13 mm. In this subgroup analysis, the automated model continued to show equal sensitivity, but increased specificity relative to early-to-late diastolic transmitral velocity ratio, e′, and strain.
Conclusions Our results suggested that machine-learning algorithms can assist in the discrimination of physiological versus pathological patterns of hypertrophic remodeling. This effort represents a step toward the development of a real-time, machine-learning–based system for automated interpretation of echocardiographic images, which may help novice readers with limited experience.
Perspectives
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: Machine-learning algorithms can automate assessment of physiological STE data with diagnostic accuracy comparable to conventional 2D Doppler echocardiography.
TRANSLATIONAL OUTLOOK: With further experience in a wider diversity of settings, automated machine-learning systems could reduce variability and improve the diagnostic accuracy of STE in routine clinical practice.
Assessing risk and avoiding complications in breast reduction requires a meticulous history, systematic physical examination, management of expectations, and careful consideration and execution of operative technique. Attention should be paid to comorbidities. Shape, symmetry, contours, scar location, skin quality, nipple-areolar complex (NAC) shape, NAC position relative to inframammary fold, and NAC position relative to the volume of the breast should be evaluated. Because complications cannot always be anticipated, informed consent is a vital part of managing expectations. Intraoperative considerations include blood pressure control, limiting tension, delayed healing and tissue loss, and using applied anatomy to avoid malposition and asymmetry.
Objective:
Machine learning methods are flexible prediction algorithms that may be more accurate than conventional regression. We compared the accuracy of different techniques for detecting clinical deterioration on the wards in a large, multicenter database.
Design:
Observational cohort study.
Setting:
Five hospitals, from November 2008 until January 2013.
Patients:
Hospitalized ward patients INTERVENTIONS:: None MEASUREMENTS AND MAIN RESULTS:: Demographic variables, laboratory values, and vital signs were utilized in a discrete-time survival analysis framework to predict the combined outcome of cardiac arrest, intensive care unit transfer, or death. Two logistic regression models (one using linear predictor terms and a second utilizing restricted cubic splines) were compared to several different machine learning methods. The models were derived in the first 60% of the data by date and then validated in the next 40%. For model derivation, each event time window was matched to a non-event window. All models were compared to each other and to the Modified Early Warning score, a commonly cited early warning score, using the area under the receiver operating characteristic curve (AUC). A total of 269,999 patients were admitted, and 424 cardiac arrests, 13,188 intensive care unit transfers, and 2,840 deaths occurred in the study. In the validation dataset, the random forest model was the most accurate model (AUC, 0.80 [95% CI, 0.80-0.80]). The logistic regression model with spline predictors was more accurate than the model utilizing linear predictors (AUC, 0.77 vs 0.74; p < 0.01), and all models were more accurate than the MEWS (AUC, 0.70 [95% CI, 0.70-0.70]).
Conclusions:
In this multicenter study, we found that several machine learning methods more accurately predicted clinical deterioration than logistic regression. Use of detection algorithms derived from these techniques may result in improved identification of critically ill patients on the wards.
Reduction mammaplasty surgery is well known to produce improvement in a wide range of symptoms associated with macromastia. Health care insurers frequently stipulate a minimum resection volume to qualify for coverage, limiting access to surgery for many. The authors aimed to identify whether small volume resections do produce symptomatic improvement, comparing preoperative and postoperative experience of symptoms across a range of tissue resection volumes.
Reduction mammaplasty patients were given a custom-designed questionnaire at routine postoperative follow-up appointments, asking them to rate their preoperative and postoperative experience of 9 symptoms related to macromastia. Results were compiled and analyzed alongside data from patient case notes. Of 661 patients identified as being eligible for inclusion in the study, 410 had sufficiently complete data to proceed to statistical analysis. Patients were divided into 6 groups based on volume of breast tissue resected. A Schnur sliding scale percentile was also calculated for all patients. Statistical analysis of preoperative symptom prevalence and postoperative symptom change was carried out. Further analysis to examine for evidence of trend in symptom improvement across groups was implemented using the Jonckheere-Terpstra test for ordered alternatives.
Patients who go on to have larger volumes of breast tissue resected were found to experience back pain, shoulder grooves, breast pain, rashes under the breast, exercise intolerance, and poor posture more frequently than those who go on to have smaller resections (P < 0.0005 for all). However, across the range of resection volumes, preoperatively symptomatic patients experienced significant improvement in several symptoms. Results suggested that a larger resection volume may correspond with greater improvement in back pain, neck pain, and poor posture.
We found that reduction mammaplasty has a positive impact on a range of symptoms, even with lower volume resections and regardless of body surface area-calculated adjustments. This adds further weight to the argument that patients should not be denied access to the surgery based on arbitrary volume restrictions. We advocate freedom for the surgeon to make a decision on potential benefits of surgery based around the needs of each individual patient.
Reduction mammaplasty (RM) is generally thought of as a reconstructive procedure, frequently but variably reimbursed by third-party payers. The purpose of this study was to assess US plastic surgeons' opinions of and interactions with the insurance coverage environment surrounding the reimbursement of RM.
The RM policies of 15 regional and nationwide health insurance carriers were analyzed. A survey regarding RM was distributed to all members of the American Society of Plastic Surgeons and subsequently analyzed.
Most insurance carriers require a minimum resection weight, a minimum age, and a conservative therapy trial. A total of 757 surgeons responded to our survey. Seventy-six percent of the respondents believe that only some RM procedures should be covered by insurance. Sixty-four percent feel that symptoms are the most important factor in the surgeon's determination of medical necessity. Fifty-seven percent state that a breast resection weight of 500 g or greater is required for coverage in their region. Seventy-one percent believe that this weight should be less than 500 g per breast. If the surgeon estimates that he/she will remove 500 g per breast, the minimum weight for coverage, 61% of the surgeons would have patients sign a statement of liability for payment. If the intraoperative resection weight is inadequate, 45.6% would not remove additional tissue, risking nonpayment; 32.7% would complete the procedure and inform the patient that payment is out-of-pocket.
Insurance reimbursement for RM varies in approval by carrier. Surgeons believe that signs and symptoms of macromastia determine medical necessity, whereas insurance carriers place a larger emphasis on resection weights.
After reading this article, the participant should be able to: 1. Accurately state the indications for breast imaging prior to breast reduction; 2. List the modifiable risk factors in a woman considering breast reduction. 3. Use perioperative antibiotics in an evidence based fashion. 4. Identify factors that are associated with higher rates of perioperative complications. 5. Describe the risks and benefits of breast infiltration with epinephrine. 6. Describe the pros and cons of using drains following breast reduction. 7. Describe the incidence of invasive breast cancer in surgical specimens compared to autopsy specimens. 8. Identify common questionnaires that can be used to track short and long-term outcomes following breast reduction. 9. List at least three current practices that are now evolving and changing based on evidence based medicine.
This paper is designed to summarize key evidence based steps in the care of women undergoing reduction mammaplasty. In addition, the authors identify gaps between how plastic surgeons practice breast reduction and what the best evidence supports. The article was prepared to accompany practice-based assessment with ongoing surgical education for the Maintenance of Certification Program of the American Board of Plastic Surgery.
In May of 2011, the Executive Committee of the American Society of Plastic Surgeons approved an evidence-based guideline on reduction mammaplasty developed by the American Society of Plastic Surgeons Health Policy Committee. The guideline addresses six clinical questions: procedural efficacy as noted by relief of symptoms, resection weight, the impact of body mass index on surgical complications, use of prophylactic antibiotics, use of drains, and effect on quality of life. The evidence indicates that resection volume is not correlated directly to the degree of postoperative symptom relief. Increased breast resection weight may increase the risks of complication. The evidence is inconclusive on whether increased body mass index is associated with increased risk of complications. Perioperative antibiotics may reduce the risk of infection associated with reduction mammaplasty, and in standard reduction mammaplasty procedures without liposuction, the use of drains is not beneficial. Reduction mammaplasty has been shown to improve quality of life.
Current guidelines used to predict appropriate resection weight for patients undergoing reduction mammaplasty are typically based on relatively nondescript patient characteristics and are most often inaccurate. The determination of patient measurements that correlate with resection weight could enable appropriate resection weight to be predicted more precisely and on an individualized basis. To better elucidate this, data from 348 patients undergoing bilateral reduction mammaplasty (696 breasts) between October 2001 and March 2009 were reviewed retrospectively. The association between resection weight and sternal notch to nipple distance (SNN), inframammary fold to nipple distance (IMFN), and body mass index (BMI) was assessed. Regression analysis demonstrated a strong correlation between resection weight and SNN distance (r = 0.672, P < 0.001), IMFN distance (r = 0.467, P < 0.001), and BMI (r = 0.510, P < 0.001). The strongest correlation was observed after incorporating all 3 parameters (r = 0.740, P < 0.001). This enabled the calculation of a formula to predict resection weight: Predicted weight = 40.0(SNN) + 24.7(IMFN) + 17.7(BMI) - 1443 In conclusion, resection weight correlates strongly with SNN, IMFN, and BMI in patients undergoing reduction mammaplasty. When considered together, resection weight can be predicted with a strong degree of accuracy.
Ninety-two of 220 plastic surgeons submitted information from 600 women (the last 15 to 20 reduction mammaplasties by each surgeon) regarding height, weight, and amount of breast tissue removed. In a second survey to estimate percentages of women who sought reduction mammaplasties for purely cosmetic reasons, for mixed reasons, and for purely medical reasons, 132 of the same 220 surgeons responded. All women whose removed breast weight was less than the 5th percentile sought the procedure for purely cosmetic reasons, and all women whose breast weight was greater than the 22nd percentile sought the procedure for medical reasons. Those women whose removed breast tissue weight was between the 5th and the 22nd percentiles had mixed reasons for requesting the procedure.
Outcome studies of the value of reduction mammaplasties have only recently appeared in the literature. Medical directors of insurance companies and managed care plans have been reluctant to pay for reduction mammaplasties, citing the uncertainty of the medical necessity of the procedure. They have defended their position by stating that the medical literature is devoid of studies documenting that reduction mammaplasty is medically beneficial to the patient. For this reason, reduction mammaplasty is often excluded from health care benefit plans. Because of the need for outcome studies for this procedure, the charts of 363 consecutive patients who had reduction mammaplasty at the Mayo Clinic from January of 1986 to December of 1993 were reviewed. Questionnaires were sent to all these patients asking them to evaluate their outcome, and 328 responded (90.4 percent response rate). Of the respondents, 94.2 percent believed that the procedure was completely or very successful, and only 1.5 percent believed that it was not very successful or completely unsuccessful. The symptoms most frequently reported by patients preoperatively were as follows: uncomfortable feeling about their body, 97.0 percent: inability to find clothes that fit, 95.7 percent; pain in bra-strap groove, 92.4 percent; shoulder pain, 86.0 percent; inability to run, 79.3 percent; upper back pain, 79.0 percent; inability to participate in sports, 77.4 percent; neck pain, 70.7 percent; lower back pain, 64.0 percent; and intertrigo, 61.0 percent. The symptoms least frequently reported by patients preoperatively were as follows: pain or numbness in the hands, 22.6 percent; headaches, 30.2 percent; arm pain, 35.4 percent; and breast pain, 58.2 percent. These symptoms were either relieved or partially relieved in 88 percent or more of the patients. Of the 328 patients, 97.3 percent responded that they definitely or probably would have the procedure again, and only 1.2 percent definitely or probably would not have the operation again. Evaluation of medical treatment used to relieve symptoms showed a marked decrease in the need for such measures after reduction mammaplasty. Study of the charges for the procedure revealed that the setting of practice parameters for the procedure and the use of an ambulatory surgery center significantly decreased the charges for the procedure. This outcome study supports the hypothesis that reduction mammaplasty is an effective procedure and the treatment of choice for symptomatic mammary hyperplasia.
Obesity and specimen weight have both been associated with a higher incidence of complications for patients undergoing reduction mammaplasty. However, obesity has been arbitrarily and inconsistently defined, and the degree of obesity has not been considered in these previous studies. Because insurance companies are increasingly demanding weight loss before authorizing surgery, the relationship of obesity and breast size to complications is of great importance. Upon critical review of the literature, a number of fundamental questions remain unanswered. If complications are more frequent in the obese patient, are these complications directly proportional to the degree of obesity? Also, if the patient is required to lose weight before surgery, is weight loss effective in reducing complication rates? In an attempt to clarify these issues, 395 patients who underwent reduction mammaplasty over a 10-year period were reviewed retrospectively. Patients were arbitrarily divided into five groups in which, depending on their degree of relative obesity, they were classified as less than 5 percent, 5 to 10 percent, 10 to 15 percent, 15 to 20 percent, or greater than 20 percent above average body weight. To evaluate the relationship of specimen weight to complications, patients were also arbitrarily divided into five groups according to weight of the breast reduction specimen, which was classified as less than 300 g, 300 to 600 g, 600 to 900 g, 900 to 1200 g, and greater than 1200 g reduction per breast. Complications were then divided into local and systemic and major and minor. When bilateral reductions alone were analyzed (n = 267), there was a statistically significant increase in complication rate in the obese (p = 0.01). However, when the obese population was further subdivided according to their degree of obesity (less than 5 percent, 5 to 10 percent, 10 to 15 percent, 15 to 20 percent, and greater than 20 percent above average body weight), no further correlation was found. However, the relationship between specimen weight per breast and complications was much stronger with a direct correlation existing between increasing specimen weight and the incidence of complications. Although this study has shown that patients who are average body weight have fewer complications than obese patients after breast reduction surgery, it has not shown an increasing incidence of complication with increasing degrees of obesity. The implications of these findings and their relationship for denying patients surgery on the basis of weight alone are discussed in detail.
Women seeking consultation for the surgical relief of symptoms associated with breast hypertrophy have been the focus of many studies. In contrast, little is known about those women with breast hypertrophy who do not seek symptomatic relief. The purpose of this study was to describe the health burden of breast hypertrophy by using a set of validated questionnaires and to compare women with breast hypertrophy who seek surgical treatment with those who do not. In addition, this latter group was compared with a group of control women without breast hypertrophy. Women seeking consultation for surgery were recruited from 14 plastic-surgery practices. Control subjects were recruited by advertisements in primary-care offices and newspapers. Women were asked to complete a self-report questionnaire that included the European Quality of Life (EuroQol) questionnaire, McGill Pain Questionnaire, Multidimensional Body Self Relations Questionnaire (MBSRQ), the Short Form-36 (SF-36) questionnaire, and questions regarding breast-related symptoms, comorbidities, and bra size. Descriptive statistics were compiled for three groups of women: (1) hypertrophy patients seeking surgical care, (2) hypertrophy control subjects (those whose reported bra-cup size was a D or larger), and (3) normal control subjects (those whose reported bra-cup size was an A, B, or C). The multiple linear regression method was used to compare the health burdens across groups while adjusting for other variables. Two hundred ninety-one women seeking surgical care and 195 control subjects were enrolled in the study. The 184 control subjects with bra-cup information available were further separated into 88 hypertrophy control subjects and 96 normal control subjects. In the control group, bra-cup size was correlated with health-burden measures, whereas in the surgical candidates, it was not. When scores were compared across the three groups, significant differences were found in all health-burden measures. The surgical candidates scored more poorly on the EuroQol utility, McGill pain rating index, MBSRQ appearance evaluation, physical component scale of the SF-36, and on breast symptoms than did the two control groups. In addition, the hypertrophy control subjects scored more poorly than the normal control subjects. With multiple linear regression analysis incorporating important potential confounders, the poorer scores in the surgical candidates remained statistically significant. It was concluded that breast hypertrophy in those seeking surgical care and those not seeking surgery has a significant impact on women's quality of life as measured by validated and widely used self-report instruments including the EuroQol, MBSRQ, McGill Pain Questionnaire, and the SF-36. Likewise, a new assessment instrument for breast-related symptoms also demonstrated greater symptomatology in women with breast hypertrophy.
Indications for breast reduction surgery include neck pain, back pain, shoulder pain, and an intertriginous rash. Previous studies have established that bilateral breast reduction surgery is highly effective in relieving these symptoms. Third-party payers may refuse to cover breast reduction surgery for obese patients. In addition, some surgeons turn down obese breast reduction candidates, perhaps fearing that they will not achieve symptom relief or that the complication rate will be prohibitive. It is common for insurers to require an arbitrary minimum volume to be resected in order for them to reimburse for the procedure. This study was conducted to see whether patients with varying levels of obesity had any difference in surgical outcomes compared with nonobese patients with regard to symptom relief and complication rate. The authors also studied the relationship between volume of tissue resected and symptom relief and complication rate.
One hundred eighty-six consecutive patients underwent bilateral breast reduction surgery performed by a single surgeon using the inferior pedicle Wise pattern technique or the free nipple graft technique. Body mass index was calculated for each patient.
Significant postoperative symptom relief occurred in 97 percent of patients. Statistical analysis demonstrated no difference among the various body mass index groups in terms of symptom relief or development of complications, nor was there any correlation between volume of tissue resected and relief of symptoms or complications.
The authors conclude that there is no justification for discriminating against obese patients in either the performance of breast reduction surgery or the provision of insurance coverage for the same. The authors find no justification for denial of insurance coverage for patients in whom lesser tissue volumes are to be resected.
Analysis of complication data derived from the Breast Reduction Assessment: Value and Outcomes (BRAVO) study, a 9-month prospective, multicenter trial, is presented.
Data derived from 179 patients were analyzed, including bivariate associations between complications and single predictor variables (Fisher's exact test or chi-square testing) or continuous variables (two-sample t test) and, finally, logistic regression.
The overall complication rate was 43 percent (77 patients). Simple, bivariate analysis linked preoperative breast volume, shoulder strap grooving, and a vertical incision with an increased incidence of complications (p < 0.05, 0.02, and 0.02, respectively). Delayed wound healing, the most common complication, correlated directly with average preoperative breast volume (p < 0.045), average resection weight/breast (p < 0.027), and smoking (p < 0.029) and inversely with age (p < 0.011). Vertical incision techniques were associated with an increased complication frequency (p < 0.05) without a link to specific complications. Logistic regression analysis associated resection weight as the sole variable for increased risk of complications (p = 0.05) and with absolute number of complications [mean resection weight of 791 g for patients without complications versus 847, 882, and 1752 g for patients with one, two, and three complications, respectively (p = 0.0022)]. Each 10-fold increase in resection weight increased the risk of complication 4.8 times and increased the risk of delayed healing 11.6 times.
Complication data revealed several significant features: (1) resection weight correlated with increased risk and absolute number of complications; (2) delayed healing correlated directly with resection weight and inversely with increasing age, anesthesia times, and preoperative Short Form-36 bodily pain score; (3) a vertical incision may be associated with increased incidence of complications but requires further analysis; and most importantly, (4) the presence of complications had no negative effect on improvement in Short Form-36 and Multidimensional Body-Self Relations Questionnaire scores.
Neck, shoulder, and lower back pain and reduction of functional capacity can be caused by breast hypertrophy. Reduction mammaplasty appears to improve these aspects. After a systematic review of the literature, no scientific evidence was found to confirm this hypothesis. The authors' objective was to evaluate the impact of reduction mammaplasty on pain and functional capacity of patients with mammary hypertrophy.
One hundred patients with mammary hypertrophy, between 18 and 55 years old, with no previous mammary surgery, were consecutively selected from the Plastic Surgery Outpatient Clinic of the Universidade Federal de São Paulo-Escola Paulista de Medicina and randomly allocated into two groups. Group A (n = 50) underwent mammaplasty reduction immediately, whereas group B patients (n = 50) were placed on a waiting list (control group). All patients were interviewed for clinical and demographic data and evaluated to measure pain and functional capacity. To measure shoulder, neck, and lower back pain, a visual analogue scale (0 = no pain, 10 = unbearable pain) was used. The Stanford Health Assessment Questionnaire (HAQ-20), Brazilian validated version (0 = best, 3 = worst), was applied to assess functional capacity. Pain and functional capacity were measured at baseline and 6 months after surgery.
Forty-six of 50 patients, from both groups, completed the study. The mean (+/-SD) patient age in groups A and B was 31.6 +/- 11 and 32.3 +/- 10 years, respectively. The mean breast tissue weight was 1052 +/- 188 g. Functional capacity in group A was improved 6 months after reduction mammaplasty, compared with group B (control), in the following aspects: getting dressed, getting up, walking, maintaining personal hygiene, reaching, and grasping objects. The mean pain intensity dropped in the lower back, from 5.7 to 1.3; in the shoulders, from 6.1 to 1.1; and in the neck, from 5.2 to 0.9.
Reduction mammaplasty improved functional capacity and relieved pain in the lower back, shoulders, and neck of patients with mammary hypertrophy.
Reduction mammaplasty has been shown to be efficacious in reducing the burden of symptoms and improving the quality of life for patients with macromastia. However, most insurance carriers will not reimburse for mammaplasties involving less than 1000 g of total tissue resected. To refute this arbitrary policy, the authors set out to examine the effect of reduction mammaplasty in which less than 1000 g of breast tissue was resected on patients' macromastia-related symptoms and macromastia-related quality-of-life factors.
All patients were given a custom-designed questionnaire designed to evaluate macromastia-related symptoms and other macromastia-related quality-of-life issues. Patients were then provided the same questionnaire at their final postoperative visit between 3 and 12 months after surgery.
A total of 59 patients underwent reduction mammaplasty of less than 1000 g. Reduction mammaplasty less than 1000 g resulted in significant decreases in all macromastia-related symptoms analyzed, including upper back pain, lower back pain, neck pain, arm pain, shoulder pain, hand pain, breast pain, headaches, rashes, and/or itching and painful bra strap grooving (all p < 0.00002). Furthermore, reduction mammaplasty resulted in significant improvement in all quality-of-life factors analyzed, including difficulty buying clothes and bras, difficulty participating in sports, and difficulty running (all p < 0.00001).
Reduction mammaplasty totaling less than 1000 g offers substantial relief of macromastia-associated symptoms and results in significant improvement in patients' quality of life. This prospective study conclusively demonstrates that reduction mammaplasty totaling less than 1000 g should be a fully reimbursable procedure.
Insurance companies evaluate the medical necessity for breast reduction surgery based on internal company medical policies, but the correlation of insurance company criteria to the scientifically established indications for reduction mammaplasty has never been studied. The authors obtained 90 insurance company medical policies for reduction mammaplasty to determine whether the criteria on which coverage determinations are made are consistent with published data regarding the indications for this procedure.
The authors reviewed the medical literature on reduction mammaplasty and identified what conclusions can reasonably be drawn from this literature on the common insurance criteria used to determine medical necessity for reduction mammaplasty. Conclusions based on the medical literature regarding volume of reduction, symptom presentation, conservative therapy, obesity, presence of bra strap grooving and intertrigo, and age at time of reduction were formulated, and these conclusions were compared with the criteria of 90 different health insurance reduction mammaplasty medical policies.
The authors were unable to identify any medical policies that could be supported in entirety by the medical literature and many that are completely unfounded based on the medical literature.
Insurance company medical policy requirements with respect to reduction mammaplasty are, in many cases, arbitrary and without scientific basis. Requirements for a specific volume of reduction, a minimum age, a maximum body weight, and a trial of conservative therapy are required by the majority of managed care medical policies, even though scientific support for any of these requirements is not evident in the medical literature.
Medical Policy: Reduction Mammaplasty for Breast-Related Symptoms
- Bluecross Blueshield
- Massachusetts
BlueCross BlueShield of Massachusetts. Medical Policy: Reduction Mammaplasty
for Breast-Related Symptoms. Boston, MA; 2020.
Involvement of machine learning tools in healthcare decision making
- Smdac Jayatilake
- G U Ganegoda
Jayatilake SMDAC, Ganegoda GU. Involvement of machine learning tools in
healthcare decision making. J Healthc Eng. 2021;2021:6679512.