Incomplete Excisions of Extremity Soft Tissue Sarcomas Are Unaffected by

Insurance Status or Distance From a Sarcoma Center

VIGNESH K. ALAMANDA, BS,1 GADINI O. DELISCA, BS,1 KRISTIN R. ARCHER, PhD, DPT,1

YANNA SONG, MS,2 HERBERT S. SCHWARTZ, MD,1 AND GINGER E. HOLT, MD1*

1Department of Orthopaedics and Rehabilitation, Vanderbilt University Medical Center, Nashville, Tennessee2Department of Biostatistics, Vanderbilt University Medical Center, Nashville, Tennessee

Background: Soft tissue sarcomas (STS) continue to be excised inappropriately without proper preoperative planning. The reasons for this remainelusive. The role of insurance status and patient distance from sarcoma center in influencing such inappropriate excisions were examined in thisstudy.Methods: This retrospective review of a single institution prospective database evaluated 400 patients treated for STS of the extremities betweenJanuary 2000 and December 2008. Two hundred fifty three patients had a primary excision while 147 patients underwent re‐excision.Wilcoxon ranksum test and either x2 or Fisher’s exact were used to compare variables. Multivariable regression analyses were used to take into account potentialconfounders and identify variables that affected excision status.Results: Tumor size, site, depth, stage, margins, and histology were significantly different between the primary excision and re‐excision groups;P< 0.05. Insurance status and patient distance from the treatment center were not statistically different between the two groups. Large and deeptumors and certain histology types predicted appropriate referral.Conclusions: Inappropriate excision of STS is not influenced by patient distance from a sarcoma center or by a patient’s insurance status. In thisstudy, tumor size, depth, and certain histology types predicted the appropriate referral of a STS to a sarcoma center.J. Surg. Oncol. 2013;108:477–480. � 2013 Wiley Periodicals, Inc.

KEY WORDS: soft tissue sarcoma; patient distance; insurance status; unplanned excision and incomplete excision of STS

INTRODUCTION

Soft tissue sarcomas (STS) are exceedingly rare with an estimatedincidence of 11,410 new cases each year [1]. This is in comparison tobenign soft tissue masses which are 300 times more common [2]. Therelatively low incidence of STS compared to their benign counterpartsresults in many STS being excised inappropriately without adequatepreoperative planning, biopsy, and imaging. The current standard of careis to re‐excise these incompletely excised tumors with wide margins toremove residual disease. Studies have also shown that poorly performedbiopsies initially can change functional outcomes and can precludesubsequent limb salvage surgery [3]. Despite these published articlesand attempts by sarcoma organizations to educate referring physicianson the merits of referral to a sarcoma center, one‐third of sarcomascontinue to be inappropriately managed [4,5].

Studies have shown no survival differences between tumors whichare excised in a planned resection versus tumors which are re‐excisedfollowing an incomplete excision [4–7]. The similar survival benefitsseen in patients undergoing re‐excisions as compared to patientsundergoing planned single excisions might play a role in diminishing theneed for appropriate referrals to sarcoma centers for definitive resectionsas the referring surgeon can incorrectly assume that the patient canalways undergo a re‐excision should the mass be malignant with nodisadvantages. However, re‐excision of an incompletely excised tumordoes not come without its drawbacks and include an additional surgery,increased emotional toils, increased rehabilitation and the increasedcomplexity of the definitive surgical procedure secondary tocomplications from the unplanned resection [7–10].

The purpose of this study was to investigate factors that may act as abarrier to appropriate referral of patients with STS to a sarcoma specialtycenter. Our initial hypothesis was that discrepancies would exist in (1)patient distance from a sarcoma center and (2) insurance status;

increased rates of inappropriate excisions were expected for patientswho live farther away from the sarcoma center and for those withoutinsurance. Increased distance from the treatment center and/or a lack ofinsurance might serve as an impediment to appropriate referral andmight prompt patients to request resection of questionable masseswithout increased travel or cost to the patient (Fig. 1).

MATERIALS AND METHODS

Following approval from our institution’s review board, weconducted a retrospective cohort study at a major sarcoma center toassess differences in patients who underwent re‐resection following anincomplete resection compared to that of patients who underwent asingle, planned excision. All patients undergoing surgical resection ofsoft tissue sarcoma, meeting our inclusion and exclusion criteria, at ourinstitution between January 2001 and December 2008 (n¼ 400) wereidentified based on a retrospective review of a prospective study andwere considered for the study. Patients were excluded if they wereyounger than 18 years of age, lacked adequate medical records or if they

Grant sponsor: REDCap (Research Electronic Data Capture) (1 UL1RR024975 NCRR/NIH)

*Correspondence to: Ginger E. Holt, MD, Department of Orthopaedics andRehabilitation, Vanderbilt UniversityMedical Center, 1215 21st Ave. South,Medical Center East, South Tower, Suite 4200 Nashville, TN 37232‐8774.Fax: þ1‐615‐343‐1028. E‐mail: ginger.e.holt@vanderbilt.edu

Received 24 June 2013; Accepted 10 August 2013

DOI 10.1002/jso.23427

Published online 4 September 2013 in Wiley Online Library(wileyonlinelibrary.com).

Journal of Surgical Oncology 2013;108:477–480

� 2013 Wiley Periodicals, Inc.

had a tumor with good prognosis/borderline malignancy such asdermatofibrosarcoma protuberans [11].

Patients were divided into two groups. The primary excisiongroup (n¼ 253) contained patients who were referred prior to anyattempt at surgery for their tumor. These patients underwent a definitiveplanned radical excision of the tumor with intraoperative pathologyconsultation to confirm both the diagnosis and that negative marginswere achieved. Many of these patients had a prior fine needleaspiration, core needle biopsy, or open biopsy to confirm thediagnosis prior to the index procedure. The second group (n¼ 147)consisted of patients who underwent re‐excision after being referredfollowing an incomplete excision by any surgeon not specifically trainedin treating sarcomas.

Patient demographics and tumor characteristics were collectedfrom a retrospective review of medical records. Patient characteristicsincluded age at time of surgery, sex, race, and insurance type (public,private, or none). Tumor characteristics consisted of size, depth(superficial or deep), site (upper or lower extremity), histology grade(low, intermediate, or high), margin status following definitivesurgery, and histologic subtype. Staging of the patients was alsocarried out per the guidelines recommended by the American JointCommittee on Cancer (AJCC) [12]. Patient distance from the sarcomacenter was calculated bymeasuring the distance, in miles as estimated byGoogle Maps©, between the sarcoma center and the patient’s five digitzip code.

Demographic and clinical variables including risk factors weresummarized and compared for patients undergoing single excision andre‐excision. Descriptive summaries of continuous variables werepresented in terms of interquartile range (IQR) while discretevariables were summarized in terms of frequencies and percentage.Wilcoxon rank sum and x2 or Fisher’s exact test were used to testdifferences across groups. In addition to univariate comparisons,multivariable regression analysis was used to take into account potentialconfounders [13]. This model examined patient distance, insurancestatus, tumor size, site, depth, grade and certain histology types asindependent risk factors of a tumor being appropriately excised.Statistical software R (version 1.11.1, www.r‐project.org) was used forall data analysis. Reported P values were two‐sided and a P‐value of<0.05 was considered to indicate statistical significance.

RESULTS

Demographics, clinical information and tumor characteristicsbetween the single excision and re‐excision group are presented inTable I. Statistically significant differences were noted between the twogroups in tumor site (P¼ 0.001), size (P< 0.001), depth (P< 0.001),AJCC stage (P< 0.001), microscopic margins (P¼ 0.001) andhistology type (P< 0.001). The groups were similarly matched in age(P¼ 0.09), sex (P¼ 0.10), race (P¼ 0.14), insurance status (P¼ 0.43),patient distance from the sarcoma center (P¼ 0.69), and tumor grade(P¼ 0.26).

There was no statistically significant difference between the singleexcision group and the re‐excision group in the distance between thepatient and the sarcoma center. The median distance for the singleexcision group was 114 miles with an IQR of 45.3–179 miles. Themedian distance for the re‐excision group was also 114 miles with anIQR of 56.5–175 miles (Table II).

Regression analysis found that after adjusting for tumor relatedcharacteristics, there was no statistical evidence that patient distancefrom a sarcoma center, insurance status, tumor site, or tumor gradefavored planned excisions more than unplanned excisions of STS. It wasnoted that size was significantly associated with a planned, primaryexcision. For every 1 cm increase in tumor size, the odds of the tumorundergoing a planned excision as opposed to an incomplete excisionincreased by a factor of 1.1 (95% confidence interval, CI: 1.09, 1.20;P< 0.001). Similarly, deeper tumors were also associated with aplanned, primary excision. Patients with deeper tumors were 2.54 timesmore likely (95% CI: 1.31, 4.94; P¼ 0.01) to undergo a plannedresection as compared to patients with superficial tumors. Analyzinghistology types revealed that when compared to patients with otherhistological subtypes, patients with liposarcoma had increased odds(odds ratio, OR: 4.85 (95% CI: 1.80, 13.1; P¼ 0.002)) of undergoing aplanned resection. Patients with malignant fibrous histiocytoma (MFH)had decreased odds of undergoing a planned resection (OR: 0.56 (95%CI: 0.33, 0.96; P¼ 0.03)).

DISCUSSION

Patients continue to present to specialty centers after undergoingincomplete STS resection elsewhere. Current clinical practice followingan incomplete excision has been to re‐excise the tumor and the tumor bedin order to remove residual disease and obtain negative margins. Recentstudies have shown good outcomes after re‐excision of these tumors [5–7]. However, re‐excision comes with significant drawbacks such asincreased tissue loss, increased risk of local recurrence, and the surgeonperforming the definitive surgery may not always be able to compensatefor the sub‐optimal resection performed initially [14]. Thus, thepreferred method of treating STS continues to be one of high vigilanceearly referral of masses appearing malignant to a specialty clinic andexcision of the tumor in a single, planned procedure [15–17]. Ourhypothesis was that patient distance from the sarcoma center and thepatient’s insurance status would affect whether patients underwentsingle excision or re‐excision of their soft tissue masses.

In our study, we found no discrepancy in patient distance from asarcoma center between the two groups of patients. Additionally,through multivariable analysis, this study showed that after accountingfor other possible confounding factors, distance from a sarcoma centerstill failed to be significantly different between the two groups. A patientpresenting with a questionable soft tissue mass who lives far from asarcoma center might prefer to have the mass resected withoutunderstanding the potential detrimental outcomes [4–6]. This mightespecially true in certain patient populations; elderly individuals whomight be dependent on others for transportation needs and patients withother similar social constraints might elect to not seek treatment initiallyat sarcoma centers [18]. Instead, as previously reported in other studies,

Fig. 1. Scatterplot of patient distances from sarcoma center.

Journal of Surgical Oncology

478 Alamanda et al.

these patients may have been thought to pursue treatment closer to theirhomes at a local hospital in familiar surroundings and with familiarphysicians [19].

Incomplete excision rates were the same in patients with (both privateand public) and without insurance in this study. Similar to patients wholive far away from the sarcoma center, patients with no insurance mightforgo treatment at a distant, specialized treatment center to minimizetravel expenses and a perceived higher cost of healthcare [20].Additionally, physicians may also attempt a single surgical resectionin lieu of a biopsy and surgery in two separate settings to minimize coststo the patient; however, this comes at the price of potentiallyjeopardizing limb salvage with subsequent re‐excision. Physicians arealso hesitant to transfer care of patients without insurance to colleaguesrealizing that it is a shift of a financial burden and uncompensated time—further increasing the likelihood of STS masses being excisedinappropriately. The results from this study also disprove the alternative

hypothesis of not having insurance; uninsured patients are expected tomore likely be referred early for single excision surgery as it remainscommon practice that complicated, uninsured cases are referred from thecommunity to tertiary centers. However, the results from this study showno difference between the group with insurance and the group without interms of referral to a sarcoma center prior to an attempt at resection.

These results indicate that patients and providers may not considerdistance from the sarcoma center or insurance status as important factorsin determining the next step in their care. Multivariable analysis wasalso conducted to see exactly what variables in our study populationwere influential in patients receiving planned resections of STS.Large, deep tumors and tumors of certain histology subtypes werenoted to favor patients getting planned resections. This is in accor-dance with studies that have demonstrated larger and deeper tumorsas being more likely to be appropriately excised [4,21]. It is thesmaller and superficial tumors that are frequently mistaken for benignmasses and excised inappropriately. These small, superficial tumors arefrequently assumed to be benign and are excised without appropriatepreoperative planning and diagnosis [4,6,22]. These tumors then have tobe re‐excised with wide margins to remove residual disease. While re‐excision may not affect survival outcomes, patients are subjected tounnecessary consequences including a repeat surgical procedure, greateremotional toils, loss of a larger amount of tissue and increased need forrehabilitation therapy [7–10].

Interestingly, multivariable analysis revealed that liposarcomas tendto be referred and excised appropriately even after accounting for otherconfounders such as size and site of the tumor. Liposarcomas are one ofthe better known and more commonly seen histological subtypes of thevarious kinds of sarcoma [16,23]. This may explain why referringphysicians are wary of lipomatous masses that clinically appear and

TABLE II. Multivariable Logistic Regression Analysis of Single ExcisionCompared to Re‐Excision

Variable OR P‐Value 2.5% 97.5%

Patient distance, miles 1.0 0.59 0.99 1.00Public ins. vs. no ins. 0.81 0.67 0.29 2.22Private ins. vs. no ins. 0.86 0.88 0.32 2.30Size, cm 1.1 <0.001 1.09 1.20Site (upper ext. vs. lower ext.) 0.78 0.14 0.46 1.34Depth (deep vs. superficial) 2.54 0.01 1.31 4.94Grade (II and III vs. I) 1.32 0.41 0.67 2.67Liposarcoma vs. others 4.85 0.002 1.80 13.1MFH vs. others 0.56 0.03 0.33 0.96

TABLE I. Patient Demographics and Tumor Characteristics

Variable Single excision (N¼ 253) Re‐excision (N¼ 147) P‐Value Total (N¼ 400)

Age, years, median (IQ range) 59 (46–71) 56 (41–69) 0.09 58 (44–70)Sex 0.10Male 142 (56.1%) 70 (47.6%) 212 (53.0%)Female 111 (43.9%) 77 (52.4%) 188 (47.0%)

Race 0.14White 218 (86.2%) 134 (91.2%) 352 (88.0%)Non‐White 35 (13.8%) 13 (8.8%) 48 (12.0%)

InsuranceNone 25 (9.9%) 9 (6.1%) 0.43 34 (8.5%)Public 92 (36.4%) 56 (38.1%) 148 (37.0%)Private 136 (53.7%) 82 (55.8%) 218 (54.5%)

Patient distance, miles, median (IQ range) 114 (45.3–179) 114 (56.5–175) 0.69 114 (49–177)Site 0.001Upper extremity 57 (22.5%) 52 (35.4%) 109 (27.3%)Lower extremity 196 (77.5%) 95 (64.6%) 291 (72.7%)

Size, cm, median (IQ range) 12 (8–19) 5 (3–8) <0.001 9 (5–16)Depth <0.001Superficial 23 (9.1%) 43 (29.3%) 66 (16.5%)Deep 230 (90.9%) 104 (70.8%) 334 (83.5%)

Grade 0.26I 89 (35.2%) 60 (40.8%) 149 (37.3%)II and III 164 (64.8%) 87 (59.2%) 251 (62.7%)

AJCC stage <0.001Stage I and II 100 (39.7%) 93 (63.7%) 193 (48.5%)Stage III and IV 152 (60.3%) 53 (36.3%) 205 (51.51%)

Microscopic margins 0.001Negative 214 (84.6%) 141 (95.9%) 355 (88.8%)Positive 39 (15.4%) 6 (4.1%) 45 (11.3%)

Histology type <0.001Others 75 (29.6%) 82 (55.8%) 157 (39.3%)Liposarcoma 85 (33.6%) 7 (4.8%) 92 (23.0%)Malignant fibrous histiocytoma (MFH) 93 (36.8%) 58 (39.5%) 151 (37.7%)

Journal of Surgical Oncology

Distance and Insurance on STS Resections 479

behave differently from typical lipomas [2]. The rarity of liposarcomasand the more common nature of lipomas coupled with a goodunderstanding of the differences between the two types can also raisesuspicion of malignant liposarcomas. Subsequently, such providers areable to seek timely referrals to tertiary treatment centers early in themanagement process. Additionally, magnetic resonance imaging isbetter able to delineate liposarcomas from lipomas (i.e., through thepresence of non‐adipose components) [24–26] and thus prompt animmediate referral prior to any attempts at resection.

Patient distance from the treatment center and the insurance status ofthe patient do not affect rates of inappropriate excisions at our institution.Discrepancies did not exist in terms of age, sex, and race between the twogroups. However, there continues to be high rates of unplanned STSexcisions with positive margins and contamination of surrounding tissuewith tumor cells. Alternative explanations for inappropriate excisionsshould be explored with a focus on increasing the awareness amonggeneral practitioners and other referring physicians on the importance ofappropriately diagnosing STS and the hazards associated with re‐excision.

STUDY LIMITATIONS

Since this study primarily explores the role of patient distance fromthe treatment center and patients’ insurance status at a single institution,it is inherently not representative of findings elsewhere. In this study,patients who had their tumors incompletely excised were referred for re‐excision of any residual disease. However, there may also be patientswho receive unplanned but complete resections of their tumor withnegative margins that are not captured in this study as these patients areeither not referred or fail to follow up with a sarcoma center. Despite thisstudy examining an institution and not a population level dataset, therelationships and conclusions postulated remain highly plausible.

ACKNOWLEDGMENTS

Neither the work nor the authors have any financial disclosures,conflicts of interest, or acknowledgements to report. The work also hadno funding including grants from the NIH.

REFERENCES

1. Siegel R, Naishadham D, Jemal A: Cancer statistics, 2013. CACancer J Clin 2013;63:11–30.

2. Weiss SW, Goldblum JR: Enzinger and Weiss’ soft tissue tumors.St. Louis, MO: Mosby Elsevier; 2008.

3. Mankin HJ, Mankin CJ, Simon MA: The hazards of the biopsy,revisited. Members of the musculoskeletal tumor society. J BoneJoint Surg Am 1996;78:656–663.

4. Alamanda VK, Crosby SN, Archer KR, et al.: Primary excisioncompared with re‐excision of extremity soft tissue sarcomas—Isanything new? J Surg Oncol 2012;105:662–667.

5. Lewis JJ, Leung D, Espat J, et al.: Effect of reresection in extremitysoft tissue sarcoma. Ann Surg 2000;231:655–663.

6. Fiore M, Casali PG, Miceli R, et al.: Prognostic effect of re‐excisionin adult soft tissue sarcoma of the extremity. Ann Surg Oncol2006;13:110–117.

7. Chandrasekar CR, Wafa H, Grimer RJ, et al.: The effect of anunplanned excision of a soft‐tissue sarcoma on prognosis. J BoneJoint Surg Br 2008;90:203–208.

8. Trovik CS: Local recurrence of soft tissue sarcoma. A ScandinavianSarcoma Group Project. Acta Orthop Scand Suppl 2001;72:1–31.

9. Giuliano AE, Eilber FR: The rationale for planned reoperation afterunplanned total excision of soft‐tissue sarcomas. J Clin Oncol1985;3:1344–1348.

10. HoshiM, IeguchiM, TakamiM, et al.: Clinical problems after initialunplanned resection of sarcoma. Jpn J Clin Oncol 2008;38:701–709.

11. Bowne WB, Antonescu CR, Leung DH, et al.: Dermatofibrosar-coma protuberans: A clinicopathologic analysis of patients treatedand followed at a single institution. Cancer 2000;88:2711–2720.

12. Edge SB: AJCC cancer staging manual. New York; London:Springer; 2009.

13. Fine JP, Gray RJ: A proportional hazards model for the subdistribu-tion of a competing risk. J Am Stat Assoc 1999;94:496–509.

14. Potter BK, Adams SC, Pitcher JD, Jr., et al.: Local recurrence ofdisease after unplanned excisions of high‐grade soft tissuesarcomas. Clin Orthop Relat Res 2008;466:3093–3100.

15. Gustafson P, Dreinhofer KE, Rydholm A: Soft tissue sarcomashould be treated at a tumor center. A comparison of qualityof surgery in 375 patients. Acta Orthop Scand 1994;65:47–50.

16. Gerner RE, Moore GE, Pickren JW: Soft tissue sarcomas. Ann Surg1975;181:803–808.

17. Zornig C, Peiper M, Schroder S: Re‐excision of soft tissuesarcoma after inadequate initial operation. Br J Surg 1995;82:278–279.

18. Patel A, Kolt GS, Keogh JW, et al.: The Green Prescription andolder adults: What do general practitioners see as barriers? J PrimHealth Care 2012;4:320–327.

19. Alvesson HM, Lindelow M, Khanthaphat B, et al.: Shapinghealthcare‐seeking processes during fatal illness in resource‐poorsettings. A study in Lao PDR. BMC Health Serv Res 2012;12:477.

20. Billmeier SE, Ayanian JZ, Zaslavsky AM, et al.: Predictors andoutcomes of limited resection for early‐stage non‐small cell lungcancer. J Natl Cancer Inst 2011;103:1621–1629.

21. Venkatesan M, Richards CJ, McCulloch TA, et al.: Inadvertentsurgical resection of soft tissue sarcomas. Eur J Surg Oncol2012;38:346–351.

22. Qureshi YA, Huddy JR,Miller JD, et al.: Unplanned excision of softtissue sarcoma results in increased rates of local recurrence despitefull further oncological treatment. Ann Surg Oncol 2012;19:871–877.

23. Pollock RE: Soft tissue sarcomas. Hamilton, ON; London: BCDecker; 2000.

24. Kransdorf MJ, Bancroft LW, Peterson JJ, et al.: Imaging of fattytumors: Distinction of lipoma and well‐differentiated liposarcoma.Radiology 2002;224:99–104.

25. Ohguri T, Aoki T, Hisaoka M, et al.: Differential diagnosis ofbenign peripheral lipoma from well‐differentiated liposarcoma onMR imaging: Is comparison of margins and internal characteristicsuseful? Am J Roentgenol 2003;180:1689–1694.

26. Drevelegas A, Pilavaki M, Chourmouzi D: Lipomatous tumors ofsoft tissue: MR appearance with histological correlation. Eur JRadiol 2004;50:257–267.

Journal of Surgical Oncology

480 Alamanda et al.