Predictors of Local Recurrence in High-grade Soft Tissue Sarcomas: Hydrogen Peroxide as a Local Adjuvant

Transcription

1 Predictors of Local Recurrence in High-grade Soft Tissue Sarcomas: Hydrogen Peroxide as a Local Adjuvant Adam N. Wooldridge, MD, MPH; Gregory P. Kolovich, MD, MPH; Martha K. Crist, RN; Joel L. Mayerson, MD; Thomas J. Scharschmidt, MD abstract Full article available online at Healio.com/Orthopedics. Search: Soft tissue sarcomas have a mortality rate of 40% to 60%, with local recurrence being a poor prognostic factor for overall survival. Three-percent nondiluted hydrogen peroxide is hypothesized to be an effective local adjuvant. The purpose of this study was to identify risk factors for local recurrence in high-grade soft tissue sarcomas and to determine whether using hydrogen peroxide as a local adjuvant reduced the risks of local recurrence and surgical-site infection. Retrospective data were collected for 106 patients surgically treated for high-grade soft tissue sarcomas between 2002 and The primary endpoint was local recurrence. Eighteen (16.98%) cases of local recurrence occurred. Predictors of local recurrence were margin status, estimated blood loss, and histology (ie, malignant peripheral nerve sheath tumor), with hazard ratios of 4.44 (95% confidence interval [CI], ), 1.19 (95% CI, ), and 9.21 (95% CI, ), respectively. Hydrogen peroxide yielded a statistically insignificant improvement in local recurrence, with a hazard ratio of 0.81 (95% CI, ) and a reduced risk of surgical site infection, with a hazard ratio of 0.52 (95% CI, ). Margin status, increased blood loss, and histologic subtype are associated with increased local recurrence risk. Using hydrogen peroxide improved local control and infection rates, but the difference was not statistically significant. The authors are from the Department of Orthopaedic Surgery and Rehabilitation (ANW), Texas Tech University Health Sciences Center, Lubbock, Texas; and the Department of Orthopaedics (GPK, MKC, JLM, TJS), The Ohio State University, Columbus, Ohio. The authors have no relevant financial relationships to disclose. This study was performed at Wexner Medical Center, The Ohio State University, Columbus, Ohio. Correspondence should be addressed to: Thomas J. Scharschmidt, MD, Division of Musculoskeletal Oncology, College of Medicine, Department of Orthopedics, The Ohio State University, 725 Prior Hall, Columbus, OH (thomas.scharschmidt@osumc.edu). doi: / e207

2 Soft tissue sarcomas are rare entities of mesenchymal origin with mortality rates of 40% to 60%. 1 In the United States in 2011, an estimated 10,980 cases of soft tissues sarcoma with an estimated 3920 deaths occurred. 2 Overall mortality is negatively affected by local recurrence. 3-6 Improvements in local control are important because local recurrence rates have been reported at 13% to 27%. 7-9 Previous studies have reported age, margin status, grade, and histology as local recurrence predictors. 3,9-10 Margin status is consistently associated with a risk of local recurrence. The application of a tumoricidal agent as a local adjuvant to the tumor bed has been hypothesized to decrease local recurrence rates. 11 In vitro studies demonstrated cytotoxic activity of hydrogen peroxide in brain 12 and breast 13 cancer. Retrospective reviews of hydrogen peroxide as a local adjuvant for giant cell tumor of bone have been effective To the current authors knowledge, no studies have reported using hydrogen peroxide as a local adjuvant in the treatment of soft tissue sarcomas. Due to the rarity of soft tissue sarcomas, ascertaining which factors improve local control is difficult. Radiation, used neoadjuvantly or adjuvantly, has reduced the local recurrence rate; however, it is associated with complications, such as infection, edema, and fibrosis. The increased usage of limb-salvage surgery instead of amputation has made local control more important. Therefore, the authors examined (1) whether additional patient and tumor variables (ie, age, tumor size, depth, location, and blood loss) affected the risk of local recurrence, and (2) whether using hydrogen peroxide as a local adjuvant reduced the local recurrence risk. In addition, the authors assessed whether using hydrogen peroxide was associated with a reduced surgicalsite infection risk. Materials and Methods This retrospective evaluation was approved by the institutional review board (protocol #2010C0011). Data were collected for 127 patients evaluated and treated by a single surgeon (T.J.S.) for high-grade soft tissue sarcoma between 2002 and One hundred six patients were included in the final analysis. Ten patients were excluded due to presentation with local recurrence and 11 patients were excluded due to inability to ascertain pre- and postoperative radiation and chemotherapy information from the medical records. All tumors in the analysis were deep (relative to the investing fascia) and high grade. Patient demographics and tumor characteristics were collected from a retrospective review. Patient characteristics included age, sex, and race. Tumor data collected included size in centimeters (for the largest diameter), location (ie, upper extremity, central, or proximal lower and distal lower extremity), and histopathology. Histological subtypes included malignant fibrous histiocytoma, leiomyosarcoma, liposarcoma, synovial sarcoma, and malignant peripheral nerve sheath tumor. Prognostic factors (ie, death, distant metastasis, and local recurrence) and the time frames in which they occurred were collected. Additional clinical factors collected were the use of pre- or postoperative chemotherapy and radiotherapy, presence of comorbid diseases (eg, diabetes mellitus and immune disorders), history of previous infections, operative time (minutes), incision size (cm), estimated blood loss (per 100-mL increase), transfusion requirement, whether resection occurred at another institution, use of hydrogen peroxide as a local adjuvant, surgical-site infections (as defined by the Centers for Disease Control), 17 and time to infection. Descriptive data are listed in Table 1. Use of hydrogen peroxide as a local adjuvant was confirmed in 79 (74.5%) of 106 patients. Median age at diagnosis was 57.5 years, and 53.8% of patients were women. Median excised tumor size was 7.25 cm, and the most common location was the proximal lower extremity (47.2%). Twelve (11.3%) instances occurred of close/positive margins, and the most common histologic subtype was malignant fibrous histiocytoma (55 [51.9%] patients). Local recurrence occurred in 18 (16.98%) patients, with a median time to local recurrence of 27 months. Moreover, 11 (10.3%) patients acquired a surgicalsite infection, whereas 10 (9.4%) patients had a wound complication (eg, wound breakdown or flap necrosis). Demographic and clinical variables were summarized and compared using Wilcoxon rank sum, chi-square, or Fisher s exact tests. Descriptive summaries of continuous variables were presented in terms of interquartile range, whereas discrete variables were summarized in terms of frequencies and percentages. Hazard ratios and 95% confidence intervals (CIs) for local recurrence were estimated with univariable and multivariable Cox proportional hazards models. 18 The multivariable model was estimated using stepwise hierarchical backward selection. The proportional hazards assumption was assessed by testing the interaction of follow-up time and testing of Schoenfeld residuals. No significant deviations occurred from the proportional hazards assumption. Analyses were conducted with Stata version 12 software (StataCorp, College State, Texas). Statistical tests were 2-sided, and a P value less than.05 indicated statistical significance. Results During follow-up (average follow-up, 7.13 years), 18 (16.98%) cases of local recurrence occurred. Although hydrogen peroxide appeared to lengthen the time to local recurrence compared with not using hydrogen peroxide (27 vs 24.2 months, respectively), the difference was not statistically significant (P5.915). The overall probability of local recurrence during follow-up is shown in Figure 1. Of the 16 variables considered in univariable analye208 ORTHOPEDICS Healio.com/Orthopedics

3 Hydrogen Peroxide in Soft Tissue Sarcomas Wooldridge et al Table 1 Patient Demographics and Tumor Characteristics Receiving Hydrogen Peroxide Variable Factor Yes (n579) No (n527) P Total (N=106) Median age (IQR), y 57 (46-70) 59 (33-73).630 a 57.5 (45-70) Sex, No. (%) Male 37 (46.8) 12 (44.4) 49 (46.2).830 b Female 42 (53.2) 15 (55.5) 57 (53.8) Race, No. (%) White 71 (89.9) 25 (92.6) 96 (90.6).99 c Other 8 (10.1) 2 (7.41) 10 (9.43) Median size (IQR), cm 7 (4.5-12) 7.5 ( ).674 a 7.25 (4.8-12) Location, No. (%) d Upper extremity 13 (16.5) 3 (11.1) 16 (15.1) Central 11 (13.9) 4 (14.8) 15 (14.2).030 c Proximal lower extremity 42 (53.2) 8 (29.6) 50 (47.2) Distal lower extremity 12 (15.2) 13 (48.2) 25 (23.6) Microscopic margin, No. (%) Negative 70 (88.6) 24 (88.9) 94 (88.7).637 c Positive 9 (11.4) 3 (11.1) 12 (11.3) Histology, No. (%) MFH 41 (51.9) 14 (51.9) 55 (51.9) Leiomyosarcoma 16 (20.3) 4 (14.8) 20 (18.9) Liposarcoma 8 (10.1) 2 (7.41).886 c 10 (9.43) Synovial sarcoma 7 (8.86) 3 (11.1) 10 (9.43) MPNST 7 (8.86) 4 (14.8) 11 (10.4) Presents with metastasis, No. (%) Yes 6 (7.59) 3 (11.1) 9 (8.49).690 c No 73 (92.4) 24 (88.9) 97 (91.5) Radiation therapy, No. (%) d No 16 (20.3) 15 (55.6) (29.2) Preoperative 1 (1.27) 1 (3.70).446 c 2 (1.89) Postoperative 63 (79.7) 11 (40.7) <.05 c 74 (69.8) Chemotherapy, No. (%) No 45 (56.96) 16 (59.3) (57.5) Preoperative 24 (30.4) 7 (25.9) (29.2) Postoperative 29 (36.7) 7 (25.9) (33.96) Excision at another hospital, No. (%) Yes 24 (34.2) 4 (5.06) 28 (26.4).135 c No 55 (69.6) 23 (85.2) 78 (73.6) Surgical type, No. (%) d Limb salvage 76 (96.2) 14 (51.8) 90 (84.9),.05 c Amputation 3 (3.80) 13 (48.1) 16 (15.1) Median follow-up, No. (IQR), mo d 31.6 ( ) 21.0 ( ).040 a 27.2 ( ) Survival status, No. (%) Alive 51 (64.6) 19 (70.4) 70 (66.0).644 c Deceased 28 (35.4) 8 (29.6) 36 (33.96) Median time to death (IQR), mo 21.2 ( ) 17 ( ).424 a 18.9 ( ) Local recurrence, No. (%) Yes 14 (17.7) 4 (14.8) 18 (17.0).99 c No 65 (82.3) 23 (85.2) 88 (83.0) Median time to local recurrence (IQR), mo 27 ( ) 24.2 ( ).915 a 27 ( ) Distant metastasis, No. (%) Yes 26 (32.9) 11 (40.7) 37 (34.9).490 c No 53 (67.1) 16 (59.3) 69 (65.1) Median time to metastasis (IQR) 12.4 ( ) 11.2 ( ).947 a 12.2 ( ) Abbreviations: IQR, interquartile range; MFH, malignant fibrous histiocytoma; MPNST, maligtnant peripheral nerve sheath tumor. a Wilcoxon rank sum test. b Chi-square test. c Fisher s exact test. d P values less than.05 indicate differences in proportion of hydrogen peroxide use for location, surgical type, and radiation therapy. e209

4 sis, margin status, postoperative metastasis, estimated blood loss, and histologic subtype were associated with a higher local recurrence risk (Figure 2; Table 2). Stepwise backward hierarchical selection yielded 3 predictors of local recurrence: margin status, estimated blood loss, and histology (Figure 3; Table 3). During follow-up, the risk of local recurrence among those with close or positive margins was 4.44 (95% CI, ) times those with negative margins. The risk of local recurrence for a 100-mL increase in estimated blood loss was 1.19 (95% CI, ) times those with no change in estimated blood loss. During the 7-year follow-up, the instantaneous risk of local recurrence among those with a malignant peripheral nerve sheath tumor was 9.21 (95% CI, ) times those with a malignant fibrous histiocytoma. Local recurrence probabilities for the follow-up period adjusted for margin status, estimated blood loss, and histology are shown in Figure 3. No difference existed in the probability of local recurrence between the hydrogen peroxide and no hydrogen peroxide groups (P5.712) (Figure 4). Regarding the authors secondary focus, 11 (10.4%) surgical-site infections occurred. The proportion of patients with surgical-site infections was smaller in the hydrogen peroxide group than the no hydrogen peroxide group (8.86% vs 14.8%, respectively), but the difference did not reach statistical significance (P5.466) (Table 3). In the univariable and multivariable analyses, only age was associated with an increased surgicalsite infection risk (Table 4). During follow-up, a 10-year increase in age correlated with a 1.56-times higher (95% CI, ) risk of local recurrence. Although statistically insignificant, the surgical-site infection risk among patients receiving hydrogen peroxide as a local adjuvant was 0.52 (95% CI, ) times higher than those receiving no local adjuvant. No association existed between operative time (hazard ratio 1.00; 95% CI, ) or incision size (hazard ratio 1.00; 95% CI, ). Although wellestablished risk factors for infection (eg, comorbid disease, previous infections, and postoperative wound complications) were associated with an increased surgicalsite infection risk, they did not reach statistical significance. Discussion Due to the strong association between local recurrence and sarcoma-related death, a better understanding of prognostic factors and how to improve local control is needed. Therefore, the authors determined which prognostic factors influence local recurrence and whether using hydrogen peroxide as 2A 2C 1 a local adjuvant reduces the risk of local recurrence and surgical-site infection. The current study had limitations. Because the sample included 106 patients with 18 local recurrences, the maximum number of predictors for a multivariable model was limited. Therefore, the probability of an increase in type II error was considerable. Hazard ratios could not be Figure 1: Kaplan-Meier curve for local recurrence without adjustment showing that the overall probability of local recurrence was 56.7% (95% confidence interval, 34.1%-74.2%) by the end of follow-up. Y-axis: survival rate. Figure 2: Kaplan-Meier curves comparing local recurrence in positive vs negative margins (A), for different histologies (B), for estimated blood loss (C), and for postoperative metastasis (D). Of the 16 variables included in univariable analysis, 4 were associated with a higher local recurrence risk: positive margins (A), histology (B), estimated blood loss (C), and postoperative metastasis (D). Y-axis: survival rate. Abbreviations: Mets, metastases; MFH, malignant fibrous histiocytoma; MPNST, malignant peripheral nerve sheath tumor; Post-op, postoperative. 2B 2D e210 ORTHOPEDICS Healio.com/Orthopedics

5 Hydrogen Peroxide in Soft Tissue Sarcomas Wooldridge et al Table 2 Secondary Outcome: Surgical-site Infection Receiving Hydrogen Peroxide Variable Yes (n579) No (n527) P Total (N5106) Infection, No. (%) Yes 7 (8.86) 4 (14.8) 11 (10.3) No 72 (91.1) 23 (85.2) 95 (89.7) Median time to infection (IQR), d 29 (14-106) 23.5 ( ).571 b 28 (14-76) Comorbid disease, No. (%) Diabetes mellitus 10 (12.6) 2 (7.41) 12 (11.3) Autoimmune disorder 6 (7.59) 3 (11.1) 9 (8.49) Median BMI (IQR), kg/m 2c 28 (24-33) 25.5 (22-29).0082 b 27 (24-32) Smoking, No. (%) Yes 12 (15.2) 4 (5.06) 16 (15.1) No 67 (84.8) 23 (85.2) 90 (84.9) Median surgery time (IQR), min 121 (78-205) 139 (81-194).612 b (80-194) Median incision size (IQR), cm c 16 (10-20) 25 (12-40).015 b 17 (12-25) Median EBL (IQR), ml 150 ( ) 100 ( ).795 b 150 ( ) Transfusion, No. (%) Yes 19 (24.1) 5 (18.5) 24 (25.4) No 60 (75.9) 22 (81.5) 82 (77.4) Wound complications, No. (%) Delayed healing 5 (6.33) 2 (7.41) 7 (6.60) Flap necrosis 2 (2.53) 1 (3.70) 3 (2.83) History of infection, No. (%) Yes 3 (3.79) 0 (0) 3 (2.83) No 76 (96.2) 27 (100) 103 (97.2) Abbreviations: BMI, body mass index; EBL, estimated blood loss; IQR, interquartile range. a Fisher s exact test. b Wilcoxon rank sum test. c P values less than.05 indicate differences in proportion of hydrogen for incision size..466 a.711 a.99 a.790 a.217 a.569 a 3 Figure 3: Adjusted Kaplan-Meier curve controlling for the variables of margin status, estimated blood loss, and histology. Y-axis: survival rate. reported for liposarcoma and synovial sarcoma due to the small sample size. Given the retrospective nature of the study, a high probability exists of selection bias with well-defined inclusion and exclusion criteria. Factors not included in the analysis were time to chemotherapy and radiotherapy or types of chemotherapy and radiation doses. Amputations and proximity to neurovascular structures were typically associated with no hydrogen peroxide in the current cohort. However, the authors could not definitely note these scenarios from the medical record without introducing recall bias. Inability to control these factors increases type II error and biases the result to the null hypothesis. Nonetheless, this is a relatively large series focusing specifically on outcomes in high-grade soft tissue sarcomas. In the current report, margin status, estimated blood loss, and histologic sub- e211

6 Table 3 Predictors of Local Recurrence Variable Univariable P HR (95% CI) Multivariable P HR (95% CI) 10-y age increase ( ) Sex ( ) Race a Size (per 1-cm increase) ( ) Margin ( ) ( ) Location Pelvic vs upper extremity ( ) Upper leg vs upper extremity ( ) Lower leg vs upper extremity ( ) Resection at another hospital ( ) Radiation Preop ( ) Postop ( ) Chemotherapy Preop (0.40, 2.90) Postop ( ) Mets At presentation ( ) Postop < ( ) Peroxide ( ) Amputation ( ) BMI ( ) 100-mL EBL increase ( ) ( ) Transfusion ( ) Histology Leiomyosarcoma vs MFH ( ) Liposarcoma vs MFH a Synovial vs MFH a MPNST vs MFH ( ) ( ) Abbreviations: BMI, body mass index; CI, confidence interval; EBL, estimated blood loss; HR, hazard ratio; mets, metastases; MFH, malignant fibrous histiocytoma; MPNST, malignant peripheral nerve sheath tumor. a An insufficient number of events occurred to calculate hazard ratios. type were predictors of local recurrence. Previous studies have reported correlations between margin status and histologic subtypes. 3,9-10,19-22 Controlling margin status is critical in preventing local recurrence. The hazard ratios observed in the current analysis are similar to those reported in the literature. 9,23 In the creation of a nomogram for the local recurrence risk reported by Cahlon et al, 9 a review of 684 patients with primary, nonmetastatic disease yielded an estimated hazard ratio of 2.37 (95% CI, ) for positive/ close margins. In a review of 278 patients comparing primary vs re-excision by Alamanda et al, 23 positive margins yielded an estimated hazard ratio of 3.76 (95% CI, ). Difficult resections near neurovascular structures increase the likelihood of close margins. Although radiotherapy can improve local control, 24 treatments are often delayed due to poor wound healing. Intraoperative local adjuvants pro- e212 ORTHOPEDICS Healio.com/Orthopedics

7 Hydrogen Peroxide in Soft Tissue Sarcomas Wooldridge et al 4 Figure 4: Graph showing the local recurrence curve for hydrogen peroxide. No statistically significant relationship existed in local recurrence between patients treated with (solid line) and without (dotted line) hydrogen peroxide. Y-axis: survival rate. vide an opportunity to improve local control. Hydrogen peroxide has been used as a local adjuvant in the control of giant cell tumor of bone Using hydrogen peroxide as a local adjuvant did not change the proportion of patients with local recurrence compared with patients receiving no local adjuvant (17.7% vs 14.8%, respectively; P5.99). Although the median time to local recurrence was longer in the hydrogen peroxide group than the no hydrogen peroxide group (27 vs 24.2 months, respectively), the difference was not statistically significant (P5.915). The hazard ratio for local recurrence with hydrogen peroxide as a local adjuvant during follow-up was 0.81 (95% CI, ). Due to the limited number of local recurrences (n54) in the pa- Table 4 Factors Predictive of Surgical-site Infection Variable Univariable P HR (95% CI) Multivariable P HR (95% CI) 10-y age increase (1.04 to 2.35) ( ) Sex (0.12 to 1.70) Race (21.84 to 2.30) Size (per 1-cm increase) (0.92 to 1.12) Margin (0.10 to 6.11) Location Pelvic vs upper extremity (0.08 to 19.55) Upper leg vs upper extremity (0.24 to 16.66) Lower leg vs upper extremity (0.21 to 19.42) Radiation a Preoperative Postoperative (0.09 to 0.99) Chemotherapy Preoperative (0.10 to 2.27) Postoperative (0.09 to 1.98) Operative time (0.9 to 1.01) Incision size (0.96 to 1.05) Amputation (0.29 to 6.24) Body mass index (0.94 to 1.15) Estimated blood loss (0.99 to 1.00) Transfusion (0.59 to 8.53) Comorbid disease Diabetes mellitus (0.39 to 8.64) Autoimmune (0.13 to 8.37) Previous infection (0.52 to 32.30) Wound quality (0.37 to 23.25) Postop wound complications a Hydrogen peroxide (0.15 to 1.81) a An insufficient number of events occurred to estimate hazard ratios. e213

8 tients who received no hydrogen peroxide (n527), conclusions cannot be drawn on the efficacy of hydrogen peroxide as a local adjuvant due to increased type II error that biases the result to the null hypothesis. Although the decreased risk of local recurrence observed with hydrogen peroxide therapy was not statistically significant, this does not preclude hydrogen peroxide as a clinically significant adjuvant. Estimated blood loss increased the risk of local recurrence. It was hypothesized that the association between estimated blood loss and local recurrence risk was due to blood transfusions. In a study on colon cancer, Amato and Pescatori 25 reported an increased risk for local recurrence if the patient received a blood transfusion. The hypothesis that blood transfusion mechanism increased the local recurrence risk is due to immunomodulation. 26 This theory is also an explanation for why local recurrence is more prominent in older patients. 27,28 For patients receiving a packed red blood cell transfusion, the local recurrence risk was 1.41 (95% CI, ) times higher than those receiving no transfusion. The association most likely did not reach statistical significance due to the sample size and collinearity with estimated blood loss. The current results are similar to those in a report of 232 patients with high-grade soft tissue sarcoma by Heslin et al 29 that reported no association between blood transfusion and local recurrence (P5.26). However, Newcomer et al 30 reported a relationship between transfusions and postoperative metastasis risk (P5.037). Several studies have determined the increased risk of local recurrence with varying histologic subtypes. In a nomogram for local recurrence, Cahlon et al 9 estimated that all histologic subtypes as compared with well-differentiated liposarcoma had 2.86 (95% CI, ) times the risk of local recurrence. Pisters et al 31 reported that fibrosarcoma had 2.5 (95% CI, ) times the risk and that malignant peripheral nerve sheath tumor had 1.8 (95% CI, ) times the risk of local recurrence. Histologic subtype and grade are vital in patient counseling and have been integrated into nomograms predicting mortality and local recurrence by the Memorial Sloan Kettering Sarcoma Registry. 32 Although hydrogen peroxide reduced the surgical-site infection risk (hazard ratio50.52) in the current study, the difference did not reach statistical significance (P5.308). Comorbid disease, previous infections, and wound quality are commonly associated with increased infection risk. Although the current results showed increased risk for these factors, they did not reach statistical significance. Due to the low incidence of surgical-site infections, a larger sample size is needed to determine whether the differences reach statistical significance. Conclusion Increased risk of local recurrence was associated with positive margin status, increased estimated blood loss, and histologic subtype (ie, malignant peripheral nerve sheath tumor), which is consistent with previous research. 3,9,10 Particular emphasis should be placed on obtaining negative margins and meticulous hemostasis to decrease local recurrence. Hydrogen peroxide reduced the risks of local recurrence and surgical-site infection, but the differences did not reach statistical significance. The authors recommend using hydrogen peroxide as a low-risk, low-cost local adjuvant. Although the current results did not reach statistical significance, hydrogen peroxide should not be ruled out as a clinically relevant local adjuvant. Further studies are needed to explore hydrogen peroxide as a local adjuvant. Appropriate validation with future randomized, controlled trials or metaanalyses is needed to determine whether hydrogen peroxide or other local adjuvants can improve local control. References 1. Gilbert NF, Cannon CP, Lin PP, Lewis VO. Soft-tissue sarcoma. J Am Acad Orthop Surg. 2009; 17(1): Siegel R, Ward E, Brawley O, Jemal A. Cancer statistics, 2011: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths. CA Cancer J Clin. 2011; 61(4): Eilber FC, Rosen G, Nelson SD, et al. Highgrade extremity soft tissue sarcomas: factors predictive of local recurrence and its effect on morbidity and mortality. Ann Surg. 2003; 237(2): Kattan MW, Leung DH, Brennan MF. Postoperative nomogram for 12-year sarcoma-specific death. J Clin Oncol. 2002; 20(3): Gronchi A, Miceli R, Fiore M, et al. Extremity soft tissue sarcoma: adding to the prognostic meaning of local failure. Ann Surg Oncol. 2007; 14(5): Abatzoglou S, Turcotte RE, Adoubali A, Isler MH, Roberge D. Local recurrence after initial multidisciplinary management of soft tissue sarcoma: is there a way out? Clin Orthop Relat Res. 2010; 468(11): Trovik CS, Bauer HC, Alvegard TA, et al. Surgical margins, local recurrence and metastasis in soft tissue sarcomas: 559 surgicallytreated patients from the Scandinavian Sarcoma Group Register. Eur J Cancer. 2000; 36(6): Karakousis CP, Proimakis C, Rao U, Velez AF, Driscoll DL. Local recurrence and survival in soft-tissue sarcomas. Ann Surg Oncol. 1996; 3(3): Cahlon O, Brennan MF, Jia X, Qin LX, Singer S, Alektiar KM. A postoperative nomogram for local recurrence risk in extremity soft tissue sarcomas after limb-sparing surgery without adjuvant radiation. Ann Surg. 2012; 255(2): Trovik CS, Bauer HC, Berlin O, et al. Local recurrence of deep-seated, high-grade, soft tissue sarcoma: 459 patients from the Scandinavian Sarcoma Group Register. Acta Orthop Scand. 2001; 72(2): Docherty JG, McGregor JR, Purdie CA, Galloway DJ, O Dwyer PJ. Efficacy of tumoricidal agents in vitro and in vivo. Br J Surg. 1995; 82(8): Mesiwala AH, Farrell L, Santiago P, Ghatan S, Silbergeld DL. The effects of hydrogen peroxide on brain and brain tumors. Surg Neurol. 2003; 59(5): Lucarotti ME, White H, Deas J, Silver IA, Leaper DJ. Antiseptic toxicity to breast carcinoma in tissue culture: an adjuvant to conservation therapy? Ann R Coll Surg Engl. 1990; 72(6): e214 ORTHOPEDICS Healio.com/Orthopedics

9 Hydrogen Peroxide in Soft Tissue Sarcomas Wooldridge et al 14. Balke M, Schremper L, Gebert C, et al. Giant cell tumor of bone: treatment and outcome of 214 cases. J Cancer Res Clin Oncol. 2008; 134(9): Ward WG Sr, Li G III. Customized treatment algorithm for giant cell tumor of bone: report of a series. Clin Orthop Relat Res. 2002; (397): Nicholson NC, Ramp WK, Kneisl JS, Kaysinger KK. Hydrogen peroxide inhibits giant cell tumor and osteoblast metabolism in vitro. Clin Orthop Relat Res. 1998; (347): Horan TC, Gaynes RP, Martone WJ, Jarvis WR, Emori TG. CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. Infect Control Hosp Epidemiol. 1992; 13(10): Cox DR. Regression models and lifetables (with discussion). J R Stat Soc B Stat Methodol ; 34: Potter BK, Adams SC, Pitcher JD Jr, Temple HT. Local recurrence of disease after unplanned excisions of high-grade soft tissue sarcomas. Clin Orthop Relat Res. 2008; 466(12): Lewis JJ, Leung D, Heslin M, Woodruff JM, Brennan MF. Association of local recurrence with subsequent survival in extremity soft tissue sarcoma. J Clin Oncol. 1997; 15(2): Heslin MJ, Woodruff J, Brennan MF. Prognostic significance of a positive microscopic margin in high-risk extremity soft tissue sarcoma: implications for management. J Clin Oncol. 1996; 14(2): Gronchi A, Lo Vullo S, Colombo C, et al. Extremity soft tissue sarcoma in a series of patients treated at a single institution: local control directly impacts survival. Ann Surg. 2010; 251(3): Alamanda VK, Crosby SN, Archer KR, Song Y, Schwartz HS, Holt GE. Primary excision compared with re-excision of extremity soft tissue sarcomas is anything new? J Surg Oncol. 2012; 105(7): McGee L, Indelicato DJ, Dagan R, et al. Long-term results following postoperative radiotherapy for soft tissue sarcomas of the extremity [published online ahead of print May 4, 2012]. Int J Radiat Oncol Biol Phys. 25. Amato A, Pescatori M. Perioperative blood transfusions for the recurrence of colorectal cancer. Cochrane Database Syst Rev. 2006; 25(1):CD Brand A. Immunological aspects of blood transfusions. Transpl Immunol. 2002; 10(2-3): Blumberg N, Heal JM. Effects of transfusion on immune function. Cancer recurrence and infection. Arch Pathol Lab Med. 1994; 18(4): Blumberg N, Triulzi DJ, Heal JM. Transfusion-induced immunomodulation and its clinical consequences. Transfus Med Rev. 1990; 4(4 suppl 1): Heslin MJ, Gaynor JJ, Newman E, et al. Effect of perioperative blood transfusion on recurrence and survival in 232 primary highgrade extremity sarcoma patients. Ann Surg Oncol. 1994; 1(3): Newcomer AE, Dylinski D, Rubin BP, et al. Prognosticators in thigh soft tissue sarcomas. J Surg Oncol ; 103(1): Pisters PW, Leung DH, Woodruff J, Shi W, Brennan MF. Analysis of prognostic factors in 1,041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol. 1996; 14(5): Memorial Sloan-Kettering Cancer Center. Soft Tissue Sarcoma Prediction Tools. org/cancer-care/adult/soft-tissue-sarcoma/prection-tools. Accessed January 15, e215