Abstract
Background
Postoperative wound complications are challenging in patients with localized extremity soft-tissue sarcomas. Various factors have been associated with wound complications, but there is no individualized predictive model to allow providers to counsel their patients and thus offer methods to mitigate the risk of complications and implement appropriate measures.
Questions/purposes
We used data from multiple centers to ask: (1) What risk factors are associated with postoperative wound complications in patients with localized soft-tissue sarcomas of the extremity? (2) Can we create a predictive nomogram that will assess the risk of wound complications in individual patients after resection for soft-tissue sarcoma?
Methods
From 2000 to 2016, 1669 patients undergoing limb-salvage resection for a localized primary or recurrent extremity soft-tissue sarcoma with at least 120 days of follow-up at eight participating United States Sarcoma Collaborative institutions were identified. Wound complications included superficial wounds with or without drainage, deep wounds with drainage because of dehiscence, and intentional opening of the wound within 120 days postoperatively. Sixteen variables were selected a priori by clinicians and statisticians as potential risk factors for wound complications. A univariate analysis was performed using Fisher’s exact tests for categorical predictors, and Wilcoxon’s rank-sum tests were used for continuous predictors. A multiple logistic regression analysis was used to train the prediction model that was used to create the nomogram. The prediction performance of the datasets was evaluated using a receiver operating curve, area under the curve, and calibration plot.
Results
After controlling for potential confounding factors such as comorbidities, functional status, albumin level, and chemotherapy use, we found that increasing age (odds ratio 1.02; 95% confidence interval, 1.00-1.03; p = 0.008), BMI (OR 1.05; 95% CI, 1.02-1.09; p = 0.004), lower-extremity location (OR 6; 95% CI, 2.87-12.69; p < 0.001), and neoadjuvant radiation (OR 2; 95% CI, 1.47-3.16; p < 0.001) were associated with postoperative wound complications (area under the curve 69.2% [range 62.8%-75.6%]).
Conclusions
We found that age, BMI, tumor location, and timing of radiation are associated with the risk of wound complications. Based on these factors, a validated nomogram has been established that can provide an individualized prediction of wound complications in patients with a resected soft-tissue sarcoma of the extremity. This may allow for proactive management with nutrition and surgical techniques, and help determine the delivery of radiation in patients with a high risk of having these complications.
Level of Evidence
Level III, therapeutic study.
Introduction
Soft-tissue sarcomas are rare neoplasms that are often treated with a multidisciplinary approach using radiation and wide local excision, with or without chemotherapy. Multimodality therapy for this disease yields excellent rates of local control and survival [2, 4-6, 20, 21]. Although there are many advantages to limb salvage, postoperative morbidity in patients with soft-tissue sarcomas may lead to functional decline, decreased quality of life, and increased number of operations to treat complications such as wound complications, which occur in 17% to 45% of patients with sarcomas [21]. Several factors have been demonstrated to contribute to postsurgical wound complications, including preoperative radiation, tumor location in the lower extremity, diabetes, tumor proximity to the skin surface, tumor size and volume, and patient age [2-6, 20, 21]. However, data on these factors are inconsistent.
There has been a strong attempt in the sarcoma community to better recognize patients who will have wound complications postoperatively. Although associated factors have been identified, because of variation in the reported factors contributing to wound complications, it is difficult to quantify whether an individual patient will have wound complications [1, 2, 7, 11, 17, 20, 21]. As such, a tool that integrates relevant prognostic factors might be useful to accurately assess the risk of postoperative wound complications.
Nomograms have been created for many malignancies. They are prognostic tools that graphically depict a statistical predictive model and are used to generate the probability of a clinical event. Nomograms address the limitations of associated factors and incorporate multiple variables to better predict clinical outcomes [15]. To our knowledge, a nomogram for postoperative wound complications in patients with soft-tissue sarcomas of the extremity does not exist.
We therefore asked: (1) What risk factors are associated with wound complications in patients with localized soft-tissue sarcomas of the extremity? (2) Can we create a predictive nomogram that will assess the risk of wound complications in individual patients after resection for soft-tissue sarcoma?
Patients and Methods
Study Design and Setting
The United States Sarcoma Collaborative is a retrospective, multi-institution database of patients with sarcomas treated between 2000 and 2016 at high-volume United States academic medical centers that treat sarcoma, including the Medical College of Wisconsin, Emory University, the University of Chicago Medicine, Stanford University, the Ohio State University, Wake Forest University, Washington University School of Medicine, and the University of Wisconsin. Data were collected retrospectively. All investigators completed training in human research and patient privacy and obtained approval for this retrospective record review from the local institutional review board. The patients’ sarcomas were categorized according to the 2009 American Joint Committee on Cancer’s staging manual, Seventh Edition [10].
Participants
All patients with localized primary or recurrent extremity soft-tissue sarcomas were identified in the collaborative database. Exclusion criteria included metastatic disease on initial presentation, age younger than 18 years, soft-tissue sarcomas in locations other than the extremities, and patients whose wound complication status was not documented. Patients who did not have follow-up of at least 120 days were also excluded because this is the timeframe in which wound complications generally occur [21]. There were 1740 patients with soft-tissue sarcomas of the extremity. Twenty-seven were excluded because of missing wound complication status and 44 were excluded because they died without a wound complication within the 120-day follow-up period. Thus, 1669 patients were included in the analysis. The frequency of missing data was reported for each analyzed variable. Missing values were rare, except for BMI (missing for 27% of patients) and albumin level (missing for 54% of patients). Descriptive and univariate analyses used an available-case approach to missing data. To generate the nonogram, we addressed missing values using mean imputation or a multiple regression analysis. There was no difference in baseline characteristics (age, tumor size, sex, and histology) between patients who had missing data and those who did not. Patients with histopathologic types demonstrating rhabdomyosarcoma, bone sarcoma, extraosseous primitive neuroectodermal tumor, Kaposi’s sarcoma, angiosarcoma, aggressive fibromatosis, and dermatofibrosarcoma protuberans were also excluded because their respective treatments, which often include surgery alone, concurrent chemotherapy and radiation, or neoadjuvant chemotherapy, differ from that of the standard treatment for soft-tissue sarcoma.
A total of 1740 patients underwent resection for a localized primary or recurrent soft-tissue sarcoma. After eliminating patients with missing values, we evaluated 1669 patients with wound complications. None of these patients were lost to follow-up in the first 120 days after their operation. The median age was 59 years, and the median tumor size was 8.5 cm. The median BMI was 27.6 kg/m2 (Table 1).
Table 1.
Patient, tumor, and treatment variables
Description of Experiment, Treatment, or Surgery
All patients underwent wide local excision for a localized primary or recurrent soft-tissue sarcoma. Wide local excision was defined as en bloc resection through normal tissue within the compartment of the tumor’s origin to obtain a layer of nontumor tissue surrounding the tumor. Patients who received radiation or chemotherapy were treated per our specific institutional guidelines.
Radiation or chemotherapy was provided at the discretion of our multidisciplinary institutional sarcoma tumor board. Preoperative or postoperative radiation was recommended when there was a concern about a higher risk of relapse, and was administered to 455 and 370 patients, respectively. Chemotherapy was given at the discretion of the medical oncologists or as part of an ongoing clinical trial.
Variables, Outcome Measures, Data Sources, and Bias
Clinical and pathologic data were collected and evaluated for all patients who met the inclusion criteria. Sixteen variables were selected a priori by clinicians and statisticians as potential factors for wound complications: age, sex, BMI, diabetes mellitus, smoking, chronic steroid use, albumin level, extremity location, tumor category (localized primary or recurrent disease), tumor size, tumor depth, neoadjuvant chemotherapy, adjuvant chemotherapy, neoadjuvant radiation, adjuvant radiation, and functional status (Table 1). A superficial wound with or without drainage, a deep wound with drainage because of dehiscence, and intentional opening of the wound within 120 days postoperatively were defined as wound complications. A patient who had one or more of these was considered to have a wound complication and patients who had none of these were not considered to have a wound complication.
Statistical Analysis, Study Size
A univariate analysis was performed for the outcome of wound complications using Fisher’s exact tests for categorical predictors and Wilcoxon’s rank-sum tests for continuous predictors. A multiple logistic regression analysis was used to train the prediction model to create each nomogram. The first 2/3 of the dataset (n = 1112; surgery dates: January 2000 to December 2010) was used to train the logistic regression model using repeated 10-fold cross-validation. Recursive feature elimination was used during the cross-validation process to select variables for inclusion in the final model. All 16 variables were considered for inclusion as well as all first-order interactions and spline terms for continuous variables. Recursive feature elimination is an automated model-selection algorithm that was chosen for its ability to handle large predictors. It removes the weakest features, removing one or a few features at a time in a loop until a specified number of features is achieved. Cross-validation is a resampling procedure that randomly partitions the training data into subsamples that are used to optimize the model selection process. It also helps avoid overfitting the model and aids in selecting the recursive feature elimination parameters.
Each model was tested on the remaining 1/3 of the dataset (n = 557; surgery dates: December 2010 to April 2016). Prediction performance in the training and test datasets was evaluated using a receiver operating characteristic curve, the area under the receiver operating characteristic curve, and a calibration plot. For all analyses, the Type I error was maintained at 0.05, and all tests were two-sided. A probability value of < 0.05 was accepted as statistically significant. All statistical analyses were performed using R version 3.4.4 (R Foundation for Statistical Computing, Vienna, Austria). The “caret” R package was used to train the prediction models, and the “rms” R package was used to create the nomogram.
Nomogram Interpretation
The resultant nomogram graphically represents a predictive model. The factors with the largest impact on postoperative wound complications are allotted 100 points. Other variables are assigned values proportional to their effect size [15, 16]. The total number of points calculated is located on the total points scale, and the corresponding value on the outcome axis delineates the probability of postoperative wound complications. The results of the multivariate analysis were incorporated into a predictive nomogram for postoperative wound complications (Fig. 1).
Fig. 1.
The nonogram for wound complications is shown.
Results
Factors Associated with Postoperative Wound Complications in Patients with Extremity Soft-tissue Sarcomas
After controlling for multiple variables such as comorbidities, functional status, albumin level, and chemotherapy use, the multivariate analysis revealed that increasing age (odds ratio 1.02; 95% confidence interval, 1.00-1.03; p = 0.008), BMI (OR 1.05; 95% CI, 1.02-1.09; p = 0.004), lower-extremity location (OR 6; 95% CI, 2.87-12.69; p < 0.001), and neoadjuvant radiation (OR 2.16; 95% CI, 1.47-3.16; p < 0.001) were associated with postoperative wound complications (area under the curve 69.2% [range 62.8%-75.6%]). Two hundred nineteen of 1669 patients (13%) had wound complications.
Predictive Nomogram Assessing the Risk of Wound Complications in Individual Patients After Sarcoma Resection
Age (adjusted OR 1.02; 95% CI, 1.00-1.03; p = 0.008), BMI (adjusted OR 1.05; 95% CI, 1.02-1.09; p = 0.004), tumor location (adjusted OR 5.7; 95% CI, 2.87-12.7; p < 0.001), and neoadjuvant radiation (adjusted OR 2.2; 95% CI, 1.47-3.16; p < 0.001) were associated with wound complications. The resultant nomogram was internally validated by the methods described (Fig. 2). The calibration plot describes how the predicted probabilities compare with the observed event rates, based on the model. The data were divided into subgroups based on the predicted probability of wound complications. For each subgroup, the observed proportion of patients with wound complications and 95% CI were calculated. The calibration plot reveals the predicted probability groups on the x-axis and the observed proportion of wound complications on the y-axis. The red one-to-one line represents the ideal calibration where the predicted probabilities agree with the observed proportion. If the observed proportion (dot) lies below the red line, the probability of wound complications predicted by the model is higher than the observed proportion. If the observed proportion (dot) lies above the red line, the probability of wound complications predicted by the model is lower than the observed proportion.
Fig. 2.
(A) This is the in-sample area under the curve and (B) corresponding calibration plot, based on the dataset used to train the model. (C) This is the outside-sample area under the curve and (D) corresponding calibration plot that reflects performance in an independent dataset.
From the validated factors described above, the nomogram can be used by combining the points for each variable and extrapolating the total number of points to a scale of the probability of wound complication. This yields the approximate risk of wound complications in any particular patient based on significant factors in the multivariate analysis (Fig. 3). Currently, this nomogram is not available publicly on the internet, but we hope to make it available on an application soon. In the meantime, the nomogram in this study can be used by others to test its usefulness in other settings.
Fig. 3.
(A) This nonogram shows the probability of wound complications in a 70-year-old man with a T4N0M0 leiomyosarcoma of the left thigh and BMI of 40 kg/m2 who underwent preoperative radiation therapy followed by surgery. (B) This nonogram shows the probability of wound complications in a 70-year-old man with a T3N0M0 leiomyosarcoma of the left upper extremity and a BMI of 40 kg/m2 who underwent preoperative radiation therapy followed by surgery.
Discussion
Postoperative wound complications are a therapeutic challenge in patients with extremity soft- tissue sarcomas, and they may prolong a patient’s hospital stay and treatment course and decrease his or her quality of life. An individualized prediction of postoperative wound complications may help with therapeutic recommendations and perioperative management to optimize care and minimize potential long-term complications in patients with soft-tissue sarcomas of the extremities. Although there are abundant data on risk factors associated with postoperative wound complications, prognostic variables have been inconsistent across published studies [1, 2, 3, 7, 11, 17, 20, 21]. Thus, there is a need to predict whether wound complications will develop in an individual patient. In this study, age, BMI, tumor location, and the timing of radiotherapy were shown to affect the development of wound complications in the multivariate analysis. These data were used to create a prognostic nomogram to predict an individual's risk of postoperative wound complications.
Limitations
The main limitation of this study was its retrospective design and inherent issues such as selection bias and missing data. The dataset was created by a consortium of academic medical centers, and there were missing values. Although missing values were uncommon, two variables important variables in the univariate analysis, BMI and albumin level, had missing values 27% and 54% of the time, respectively. Although variables were missing in the dataset, we used an available-case approach to missing data in the descriptive and univariate analyses. Missing values were addressed using the mean imputation for the multiple regression analysis used to generate the nomogram and helped account for the unrecorded variables. Lower-extremity tumors were associated with postoperative wound complications; however, there was no delineation of the anatomic compartment in the retrospective database. As such, patients who had proximal lower-extremity soft-tissue sarcomas may have a much higher probability of having postoperative wound complications than patients with soft-tissue sarcomas in other areas, and if these data had been present, they could have been considered in the analyses and prognostic nomogram. In addition, the patient population was heterogeneous in terms of histology, adjunctive treatments, and surgical factors. Additional variables that may affect wound complications such as the radiation method (intensity-modulated radiation therapy, 3D conformal radiation, or brachytherapy), surgical technique, flap closure, drain output, operation time, skin comorbidities, and anemia were not available in the database and thus not analyzed. Thus, after considering the aforementioned factors, the risk of postoperative wound complications may be adjusted. There was no loss to follow-up for the primary endpoint of wound complications because we did not include any patients with fewer than 120 days of follow-up in the study. Lastly, although these data have been validated internally, external validation may be necessary to implement this nomogram in daily practice.
Factors Associated with Postoperative Wound Complications in Patients with Extremity Soft-tissue Sarcomas
Postoperative wound complications are a persistent problem in the management of localized soft- tissue sarcomas, particularly in patients undergoing radiation. Many risk factors have been associated with this complication; however, in this study, the delivery of neoadjuvant radiation and lower-extremity tumor location had the most impact on the development of postoperative wound complications. This is consistent with the result of studies that have examined the risk factors of wound complications in patients with soft-tissue sarcomas [9, 11, 13, 19]. One study found that patients who received radiation before surgery were more likely to have wound complications than those who received radiation after surgery: 35% versus 17% [21]. It is also well-documented that lower-extremity tumors, particularly proximal lower-extremity soft-tissue sarcomas, are associated with a higher risk of wound complications than upper-extremity sarcomas [17, 19, 21]. One study found the difference to be substantial (32% in the lower extremity compared with 3% in the upper extremity) for those undergoing preoperative or postoperative radiation [21]. Moreover, proximal lower-extremity tumors comprised 84% of wound reoperations, most commonly in the anterior thigh compartment (51%) [17]. Age and BMI have not consistently been shown to affect wound outcomes [1, 2, 3, 14]. Bujko et al. [7] assessed 202 patients with extremity, torso, and head and neck soft-tissue sarcomas who underwent preoperative radiotherapy. In addition to lower-extremity tumors and radiation boost postoperatively, increasing age was associated with postoperative wound complications after the authors controlled for age, sex, BMI, diabetes mellitus, smoking, chronic steroid use, albumin level, extremity location, tumor category (localized primary or recurrent disease), tumor size, tumor depth, neoadjuvant chemotherapy, adjuvant chemotherapy, neoadjuvant radiation, adjuvant radiation, and functional status. A further study found no difference in survival or wound outcomes between patients with obesity and those without [1], but another study found that wound complications were more prevalent in patients with a low BMI than in those with a higher BMI, and arterial thrombus events occurred more frequently in patients with a high BMI than in those with a lower BMI [14].
Although studies have shown that tumor location, age, BMI, and neoadjuvant radiation affect the development of postoperative wound complications, no study has created a predictive nomogram that allows for an assessment of an individual’s risk. The present study considers all of these factors to give the treating physician the opportunity to discuss a patient’s risk with them and make decisions together regarding the patient’s care. In particular, the tumor location (adjusted OR 5.62) and preoperative radiation (adjusted OR 2.16) had the largest impact on the development of postoperative wound complications. Thus, if a patient’s risk of postoperative wound complication is excessively high, physicians could recommend postoperative radiation to decrease this risk or discuss wound closure techniques with plastic surgeons, such as the use of vascularized tissues either by rotational or free-vascularized soft-tissue transfers.
Predictive Nomogram Assessing the Risk of Wound Complications in Individual Patients After Sarcoma Resection
Nomograms can help estimate survival outcomes for several types of malignancies [8, 15, 16, 22]. For soft-tissue sarcomas specifically, they have been created and validated to predict survival and distant metastasis by identifying prognostic factors such as age and tumor size, grade, and histology [8]. Although many nomograms predict survival and disease outcomes, few have addressed the probability of and risk factors for the development of complications [12, 15, 18. Lagarde et al. [18] constructed a nomogram that predicts postoperative complications in patients with esophageal cancer who were treated with esophagectomy. Advanced age; history of stroke, transient ischemic attack, or myocardial infarction; low preoperative forced expiratory volume 1; ECG changes; and more extensive operations predicted increased complications after esophagectomy. This nomogram was externally validated by independent, separate data [12]. Although nomograms addressing toxicity are sparse, the importance of estimating an individual’s risk of complications is warranted to provide a realistic expectation for each patient and optimize proactive measures to manage toxicities. The nomogram in this study may be useful to warn healthcare providers and patients of the risk of wound complications (Fig. 3); however, these data are preliminary and external validation is warranted to provide confirmation for routine use.
Conclusions
Wound complications after soft-tissue sarcoma resection are a frequent occurrence that adversely affects a patient’s recovery and return to function. In the current study, age, BMI, tumor location, and the timing of radiation were associated with the development of postoperative wound complications. If providers are able to predict which patients are more likely to acquire a postoperative wound complication, then methods to avoid these complications and proactive management before or immediately after surgery may be used. The present study preliminarily revealed that a nomogram may help musculoskeletal oncologists identify patients at risk and take precautions such as optimizing the patient’s nutrition, including weight management; implementing flaps; or using altered radiation techniques such as intensity-modulated radiation therapy to mitigate the likelihood of these complications. This internally validated nomogram may provide an individualized prediction of postoperative wound complications in patients with resected soft-tissue sarcomas of the extremity. It also may help clinicians counsel patients, inform them of their risk, and assist these patients in modifying these factors, specifically weight loss. In addition, physicians may consider delivering radiation in the postoperative setting if the risk of wound complications is high. Identifying these risk factors may allow surgeons to optimize perioperative management in patients in whom there is a high risk that a wound complication will develop postoperatively. We recognize that other factors may contribute to wound complications, and further delineating other risk factors to be added to the nomogram are likely best identified in a prospective, controlled trial.
Acknowledgements
We thank our colleagues of the United States Sarcoma Collaborative who participated in formulating this retrospective database.
Footnotes
Each author certifies that neither he or she, nor any member of his or her immediate family, has funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.
Each author certifies that his or her institution approved the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research.
This work was performed at the Medical College of Wisconsin, Milwaukee, WI, USA.
References
- 1.Alamanda VK, Moore DC, Song Y, Schwartz HS, Holt GE. Obesity does not affect survival outcomes in extremity soft tissue sarcoma. Clin Orthop Relat Res. 2014;472:2799-2806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Arbeit JM, Hilaris BS, Brennan MF. Wound complications in the multimodality treatment of extremity and superficial truncal sarcomas. J Clin Oncol. 1987;5:480-488. [DOI] [PubMed] [Google Scholar]
- 3.Baldini EH, Lapidus MR, Wang Q, Manola J, Orgill DP, Pomahac B, Marcus KJ, Bertagnolli MM, Devlin PM, George S, Abraham J, Ferrone ML, Ready JE, Raut CP. Predictors for major wound complications following preoperative radiotherapy and surgery for soft-tissue sarcoma of the extremities and trunk: importance of tumor proximity to skin surface. Ann Surg Oncol. 2013;20:1494-1499. [DOI] [PubMed] [Google Scholar]
- 4.Barkley HT, Jr, Martin RG, Romsdahl MM, Lindberg R, Zagars GK. Treatment of soft tissue sarcomas by preoperative irradiation and conservative surgical resection. Int J Radiat Oncol Biol Phys . 1988;14:693-699. [DOI] [PubMed] [Google Scholar]
- 5.Bedi M, King DM, Shivakoti M, Wang T, Zambrano EV, Charlson J, Hackbarth D, Neilson J, Whitfield R, Wang D. Prognostic variables in patients with primary soft tissue sarcoma of the extremity and trunk treated with neoadjuvant radiotherapy or neoadjuvant sequential chemoradiotherapy. Radiat Oncol. 2013;8:60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Brant TA, Parsons JT, Marcus RB, Jr, Spanier SS, Heare TC, van der Griend RA, Enneking WF, Million RR. Preoperative irradiation for soft tissue sarcomas of the trunk and extremities in adults. Int J Radiat Oncol Biol Phys. 1990;19:899-906. [DOI] [PubMed] [Google Scholar]
- 7.Bujko K, Suit HD, Springfield DS, Convery K. Wound healing after preoperative radiation for sarcoma of soft tissues. Surg Gynecol Obstet . 1993;176:124-134. [PubMed] [Google Scholar]
- 8.Callegaro D, Miceli R, Bonvalot S, Ferguson P, Strauss DC, Levy A, Griffin A, Hayes AJ, Stacchiotti S, Pechoux CL, Smith MJ, Fiore M, Dei Tos AP, Smith HG, Mariani L, Wunder JS, Pollock RE, Sasali PG, Gronchi A. Development and external validation of two nomograms to predict overall survival and occurrence of distant metastases in adults after surgical resection of localised soft-tissue sarcomas of the extremities: a retrospective analysis. Lancet Oncol . 2016;17:671-680. [DOI] [PubMed] [Google Scholar]
- 9.Cannon CP, Ballo MT, Zagars GK, Mirza AN, Lin PP, Lewis VO, Yasko AW, Benjamin RS, Pisters PW. Complications of combined modality treatment of primary lower extremity soft-tissue sarcomas. Cancer . 2006;107:2455-2461. [DOI] [PubMed] [Google Scholar]
- 10.Edge S, Byrd DR, Compton CC, Fritz AG, Greene FL. AJCC Cancer Staging Manual. 7th ed. Chicago, IL: Springer; 2011. [Google Scholar]
- 11.Geller DS, Hornicek HJ, Makin HJ, Raskin KA. Soft tissue sarcoma resection volume associated with wound-healing complications. Clin Orthop Relat Res. 2007;459:182-185. [DOI] [PubMed] [Google Scholar]
- 12.Grotenhuis BA, Hagen P, Reitsman B, Lagarde SM, Wijnhoven BP, van Berge Henegouwen MI, Tilanus HW, van Lanschot JJ. Validation of a nomogram predicting complications after esophagectomy for cancer. Ann Thorac Surg . 2010;90:920-925. [DOI] [PubMed] [Google Scholar]
- 13.Haas RL, DeLaney TF, O’Sullivan B, Keus RB, Le Pechoux C, Olmi P, Poulsen JP, Seddon B, Wang D. Radiotherapy for management of soft tissue sarcomas: why, when, and where? Int J Radiat Oncol Biol Phys. 2012;84:572-580. [DOI] [PubMed] [Google Scholar]
- 14.Hingorani P, Seidel K, Krailo M, Mascarenhas L, Meyers P, Marina N, Conrad EU, Hawkins DS. Body mass index (BMI) at diagnosis is associated with surgical wound complications in patients with localized osteosarcoma: a report from the Children’s Oncology Group. Pediatr Blood Cancer. 2011;57:939–942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Iasonos A, Schrag D, Ganesh VR, Panageas KS. How to build and interpret a nomogram for cancer prognosis. J Clin Oncol. 2008;26:1364-1370. [DOI] [PubMed] [Google Scholar]
- 16.Kent MS, Mandrekar SJ, Landreneau R, Nichols F, Foster NR, DiPetrillo TA, Meyers B, Heron DE, Jones DR, Tan AD, Starnes S, Putnam JB., Jr A nomogram to predict recurrence and survival of high-risk patients undergoing sublobar resection for lung cancer: an analysis of a multicenter prospective study. Ann Thorac Surg . 2016;102:239-246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Korah MP, Deyrup AT, Monson DK, Oskouei SV, Weiss SW, Landry J, Godette KD. Anatomic tumor location influences the success of contemporary limb-sparing surgery and radiation among adults with soft tissue sarcomas of the extremities. Int J Radiat Oncol Biol Phys. 2012;82:933-939. [DOI] [PubMed] [Google Scholar]
- 18.Lagarde SM, Reitsman JB, Maris AK, van Berge Henegouwen MI, Busch OR, Obertop H, Zwinderman AH, van Lanschott JJ. Preoperative prediction of the occurrence and severity of complications after esophagectomy for cancer with use of a nomogram. Ann Thorac Surg . 2008;85:1938-1945. [DOI] [PubMed] [Google Scholar]
- 19.Moore J, Isler M, Barry J, Mottard S. Major wound complication risk factors following sarcoma resection. Eur J Surg Oncol. 2014;12:1671-1676. [DOI] [PubMed] [Google Scholar]
- 20.Nielsen OS, Cummings B, O'Sullivan B, Catton C, Bell RS, Fornasier VL. Preoperative and postoperative irradiation of soft tissue sarcomas: effect of radiation field size. Int J Radiat Oncol Biol Phys . 1991;21:1595-1599. [DOI] [PubMed] [Google Scholar]
- 21.O'Sullivan B, Davis AM, Turcotte R, Bell R, Catton C, Chabot P, Wunder J, Kandel R, Goddard K, Sadura A, Pater J, Zee B. Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the limbs: a randomised trial. Lancet. 2002;359:2235-2241. [DOI] [PubMed] [Google Scholar]
- 22.Stephenson AJ, Scardino PT, Eastham JA, Bianco FJ, Jr, Dotan ZA, DiBlasio CJ, Reuther A, Klein EA, Kattan MW. Postoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Clin Oncol. 2005;23:7005-7012. [DOI] [PMC free article] [PubMed] [Google Scholar]