| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Article |
From the Departments of Plastic Surgery (JYK, AY, VS, BAR, GLR, DWC) and Surgical Oncology (REP), The University of Texas M.D. Anderson Cancer Center, Houston, Texas.
ABSTRACT
Background: Treatment for soft tissue sarcoma of the upper extremity has evolved to include limb salvage techniques. We reviewed our experience with limb salvage therapy for upper extremity sarcomas with an emphasis on functional outcomes following the reconstructive surgery.
Methods: A retrospective analysis was performed of 81 consecutive patients with soft tissue sarcoma of the upper extremity who had limb salvage therapy with reconstruction by a plastic surgeon. Univariate and multivariate regression analyses of relevant outcome variables were performed. Enneking functional scores were obtained from 43 patients.
Results: The study included 16 microvascular reconstructions and 67 non-microvascular reconstructions for a total of 83 reconstructions in 81 patients. The mean defect size was 129 cm2 (standard deviation: 106 cm2). The mean total functional score was 23.1 (range, 9 to 30). Any reconstruction-related complication and preoperative chemotherapy use were associated with a 7.3 point (P = .03) and 4.7 point (P = .01) decrease in total functional score, respectively. Kaplan-Meier product-limit analysis showed 82% 5-year overall survival and 67% 5-year disease-free survival rates.
Conclusions: For soft tissue sarcoma of the upper extremity, limb salvage with good functional outcome is possible with a judicious approach to reconstruction.
Key Words: Upper extremity • Reconstruction • Soft tissue sarcoma • Outcomes
Approximately 15% of sarcomas arise in the upper extremities.1–4 Contemporary treatment strategies have focused on limb salvage, with amputation reserved for prohibitively advanced cases. Critical to the overall success of limb salvage—in terms of both oncologic outcome and functional outcome—is the integration of surgery and adjunct therapies such as chemotherapy and radiation.5,6 This multimodal approach has led to improvements in disease-free survival with limb salvage techniques.2,5,7–12 This success is predicated, however, in part, on an ability to provide durable and efficacious wound coverage following extirpation surgery. To this end, the type of coverage is an essential facet of the overall management.
Soft tissue sarcomas of the upper extremities can pose difficult oncologic and reconstructive challenges. The efficacy of limb salvage itself has been demonstrated in select series of patients.4,13,14 The feasibility of limb salvage is determined partly by the ability to reconstruct the soft tissue defect created by tumor resection. In the upper extremities, few local tissue flap options are available to repair significant defects. Especially for larger defects, the reconstructive trend has been toward microvascular flap reconstruction. With modern protocols that include aggressive chemoradiation, robust, well-vascularized tissue is paramount for expeditious healing to allow prompt initiation of adjuvant therapy, which can improve the chances of survival.13–20
To better establish the effect of reconstructive technique on functional and oncologic outcomes in patients with upper extremity soft tissue sarcoma treated with limb salvage techniques, we analyzed a single institution’s experience over an 11-year span.
PATIENTS AND METHODS
Medical records of all patients with soft tissue sarcoma in an upper extremity treated with limb salvage therapy that required the services of a plastic surgeon for reconstruction during the period 1990–2000 were reviewed for data on demographics, surgery, adjuvant therapy, and oncologic and functional outcome. We collected additional data on functional status and oncologic status by direct patient examination, interviews, and questionnaires. All data collected were entered into a database for analysis.
The Enneking scoring system was used to assess function (Table 1). Categories within this system include pain, function, emotional acceptance, hand positioning, strength, and manual dexterity. Each category is rated on a scale of 0 to 5, with 5 representing normalcy or full function.
|
For statistical analysis, the patients were grouped by type of reconstruction, microvascular or non-microvascular. Differences between groups were assessed using Student t test, Fisher exact test, or 
2 analysis, as applicable. Multiple linear regression was then used to examine the relationship between total functional score and type of flap while controlling for potential confounding factors (e.g., age, defect size, tumor location, and neoadjuvant and adjuvant therapies). A regression curve fitting method, Lowess smoothing, was used to examine the relationship between total functional score and time since reconstruction. Odds ratios of any flap complication by type of flap and patient, tumor, and treatment characteristics were examined with univariate logistic regression. The Kaplan-Meier product limit method was used to estimate and graph overall and disease-free survival curves. The log-rank test was then used to compare the survival curves. All tests were two-sided, and P <.05 were considered statistically significant. Data analyses were performed using Stata 6.0 software (Stata Corporation, College Station, TX).
RESULTS
Patients and Adjuvant Therapy
The study included 16 microvascular reconstructions and 67 non-microvascular reconstructions for a total of 83 reconstructions in 81 patients. Patient, tumor, and treatment characteristics are summarized in Table 2. The male:female distribution was 1.7:1. The mean age of the patients at surgery was 53.6 years (range, 9 to 94 years). Malignant fibrous histiocytoma was the most common histologic subtype, occurring in 34 cases (42%) (Table 3). At presentation, 62% of tumors were recurrent sarcomas and 38% were primary tumors.
|
|
Neoadjuvant chemotherapy was used in 43% of cases, and adjuvant chemotherapy was used in 34% of cases. Both were used in 26% of cases, and neither was used in 49% of cases.
Of patients, 56% received preoperative radiation therapy. The median dose was 50 Gy (range, 30 to 66 Gy); 23% received either postoperative external-beam radiation therapy or brachytherapy; and 10% received both neoadjuvant and adjuvant radiation therapy.
Reconstruction
For the microvascular reconstructions, rectus abdominis muscle or myocutaneous flaps were most commonly used (n = 7, 44%), followed by latissimus dorsi muscle or myocutaneous flaps (n = 3, 19%)(Table 4).
|
For non-microvascular reconstructions, four defects (6%) were repaired with local flaps, seven defects (10%) were repaired by complex closure, 11 (16%) required only skin grafts, and 45 (67%) necessitated pedicled flaps. Latissimus dorsi muscle, myocutaneous flap, or both were the most commonly used pedicled flaps
The overall mean defect size was 129 cm2 (SD, 106 cm2). The mean defect size in the microvascular reconstruction group was 278 cm2, compared with 112 cm2 in the non-microvascular reconstruction group (P = .01).
Tumor involvement often extended beyond soft tissue to involve nerves, blood vessels, and tendons. In eight cases (10%), vessel resection was required; nerve resection was needed in 17 cases (20%). Prostheses were placed in the elbows in four cases (25%) of microvascular reconstruction. Prostheses were placed in the elbow, humerus, or shoulder joint in nine cases (13%) of non-microvascular reconstruction.
Overall reconstruction-associated complication rate was 8%, with no significant difference between microvascular reconstructions (19%) and non-microvascular reconstructions (6%). In terms of major complications, two vessel thromboses occurred in the microvascular reconstruction group, resulting in partial flap losses and one partial flap loss in the non-microvascular reconstruction group. Univariate logistic regression analyses demonstrated no evidence of increased complications based on type of repair, age, ethnicity, sex, smoking status, tumor location, or use of radiation or chemotherapy (Table 5). Moreover, the oncologic status at presentation—recurrent or primary tumor—did not affect the complication rate (P = .28).
|
Oncologic Outcomes
Perioperative mortality rate was 0%. A total of 20 patients (25%) had local recurrences, with 2 being candidates for re-excision and coverage; 10 patients (12%) developed distant metastases alone; 2 patients (2%) developed concurrent local and distant metastases; 57 patients (70%) were disease-free at last follow-up. Kaplan-Meier product-limit analysis showed 82% 5-year overall survival and 67% 5-year disease-free survival rates. Patients in the non-microvascular reconstruction group had a significantly greater disease-free survival rate at 5 years compared with patients in the microvascular reconstruction group (75% vs. 37%, respectively; P = .01) (Fig. 1). No significant difference was seen in the 5-year overall survival rates between the non-microvascular and microvascular reconstruction groups (87% and 56%, respectively; P = .10)(Fig. 2).
|
|
Functional Outcomes
Functional Enneking data were obtained from 43 (53%) of the 81 patients in the study. Of those who responded, 11 patients had microvascular reconstructions and 32 patients had non-microvascular reconstructions. No differences were found between those who had functional data and those who did not, except for the percentage of female patients (49% and 25%, respectively; P = .04).
The mean total functional score was 23.1 (range, 9 to 30). Table 6 shows the total and component scores by treatment factors. The mean total Enneking score was significantly higher in the non-microvascular reconstruction group (24.1 vs. 20.1; P < .05). Mean scores for strength, hand positioning, function, and pain did not differ significantly among the reconstruction cohorts; however, the non-microvascular reconstruction group did have higher manual dexterity and emotional acceptance scores.
|
Also, patients who had preoperative chemotherapy had a significantly lower mean total functional score than those who did not (24.1 vs. 19.6; P < .05). In fact, this pattern held for every function component score except emotional acceptance.
Within the patient sample who had Enneking scores available, no significant differences were seen between the microvascular and non-microvascular reconstruction cohorts in terms of demographic and treatment factors except for a higher percentage of smokers in the microvascular reconstruction group (46% vs. 6%; P = .01).
The Lowess curve smoothing analysis suggested that the total functional score in patients receiving microvascular reconstructions was higher among those interviewed later than among those interviewed sooner after reconstruction and was approaching that in the non-microvascular reconstruction group at 5 years (Fig. 3).
|
Multiple linear regression analyses demonstrated that any reconstruction-related complication resulted in a reduction in total Enneking score by 7.3 points (P = .03) and that the use of preoperative chemotherapy resulted in a 4.7 point reduction (P = .01) (Table 7).
|
DISCUSSION
The type of reconstruction was partly a function of the size of the postablative defect, as the mean defect size for the microvascular reconstructions was 278 cm2 versus 112 cm2 for the non-microvascular reconstructions. The relatively large size of the defects in this study may be a function of the aggressive nature of the lesions that required reconstruction. Keeping in mind that the patients were referred to a dedicated cancer institution, it is not surprising that 62% of tumors were recurrent at presentation. This is significantly higher than in the general population of patients with upper extremity sarcomas, of which approximately 25% are recurrent at presentation.4
Most of the patients in our study received preoperative radiation therapy, with a minority receiving brachytherapy or external-beam radiation postoperatively for surgical margin control. Chemotherapy was used in the neoadjuvant or adjuvant setting in most patients. Use of these therapies did not seem to result in a higher complication rate. Univariate logistic regression analyses demonstrated that the odds ratios for reconstruction-related complications did not vary with preoperative or postoperative chemotherapy or radiation.
Overall reconstruction-related complication rate was 8%, with major complications limited to partial flap losses associated with vessel thromboses in the microvascular reconstruction group and one case of partial flap loss in the non-microvascular reconstruction group. No significant difference was seen in the complication rate between the two cohorts, despite the increased technical complexity associated with microvascular reconstructions. Interestingly, radiation therapy was not associated with an increased complication rate, although this may have been a result of the small sample size and overall low complication rates.
A higher percentage of smokers were in the microvascular reconstruction group than in the non-microvascular reconstruction group (P = .01). The tenuous nature of local or pedicled flaps in active smokers undoubtedly contributed to the selection of microvascular reconstruction for them.
In terms of oncologic outcomes, the relatively high local recurrence rate (25%) may have been resulted, in part, from the high proportion of patients who presented with local recurrences. Bray et al.14 noted a similar rate of repeated recurrence in their series: 22% versus 7% for primary excisions. Pisters et al.2 have shown that recurrence is itself a prognostic factor for a repeat recurrence. In our study, 70% of patients were disease-free at last follow-up, which is comparable to the 68% noted by Lohman et al.4 in their cohort of patients with upper extremity sarcoma who had direct closure without reconstruction.
A breakdown of survival by reconstructive technique showed no significant difference in the 5-year overall survival rates between the non-microvascular and microvascular reconstruction groups. This suggests that, although microvascular reconstruction is often used for larger defects, it also enables the ablative surgeon to use larger surgical margins and, thus, it improves the likelihood of achieving disease-free margins. Microscopically positive margins have been demonstrated to be an independent prognostic factor for recurrence and tumor-related mortality.2
Functional outcomes were available for 53% of the patients, and the mean total functional score was 23.1. The Enneking functional scoring system has been demonstrated to be useful in assessing function after soft tissue reconstruction of the extremities, with low interobserver variability and high concordance among providers and patients.13
The mean total functional score was higher in the non-microvascular reconstruction group (24.1) than in the microvascular reconstruction group (20.1; P < .05). Mean scores for strength, hand positioning, function, and pain did not differ significantly between the reconstruction cohorts; however, the non-microvascular reconstruction group did have higher manual dexterity and emotional acceptance scores. The dexterity difference can be reasonably extrapolated from the more significant ablation associated with the use of free flaps. Emotional acceptance can similarly be explained by the more extensive surgery and rehabilitation required with microvascular reconstruction than with simpler reconstructions such as local flaps or skin grafts.
Preoperative radiation therapy did not result in a significant difference in total functional score. Patients who did not have preoperative chemotherapy, however, had a significantly higher mean total functional score than did those who received preoperative chemotherapy. In fact, this pattern held for every function component score except emotional acceptance. This pattern is likely related to the more advanced, aggressive nature of the lesions requiring neoadjuvant therapy.
Multiple linear regression analyses of function-associated variables revealed that the presence of any reconstruction-related complication resulted in a 7.3 point decrease in total functional score (P = .03). This significant drop attests to the profound physiologic and psychologic effect that complications have on limb use. Preoperative chemotherapy—probably for the same reason stated above—resulted in a 4.7 point decrease in total functional score compared with no preoperative chemotherapy (P = .01).
Overall, our findings support the general conclusion that limb salvage with reconstruction in patients with upper extremity sarcoma is tolerated well. Microvascular reconstruction was associated with reduced function, and dexterity and emotional acceptance appeared to be less in this group. Pain, strength, and hand positioning did not differ, however, from those in the non-microvascular reconstruction group, despite the increased magnitude and complexity of the presenting defect in the microvascular reconstruction group. Function is, however, highly vulnerable to perioperative complications. More than the type of reconstruction, the fact that a complication occurred seemed to have an impact on functional outcome in a fashion heretofore undescribed. The effect of preoperative chemotherapy on function appeared to result from the debilitating nature of more advanced cancers. Whether some other influence on postoperative functional outcome occurred independent of the selection bias is difficult to ascertain from our data.
A suggestion is made that with longer follow-up intervals, the function improved. Because the functional points used in the Lowess curves were not sequentially obtained from the same patients over time, we cannot assess whether an actual improvement in function occurred with time or whether this is merely an epiphenomenon of patients with improved survival having improved function by virtue of the less aggressive nature of their disease.
A sequential assessment of function over time may help elucidate the relationship between survival and function. Do longer-term survivors indeed have improved function at the outset, or is the improved function a sequela of enhanced rehabilitation? A prospective study that establishes preintervention functional status and follows function through the course of surgical and adjuvant treatment and beyond would be particularly illuminating in this regard.
CONCLUSIONS
In conclusion, our study demonstrates that good functional outcome is possible in patients with upper extremity soft tissue sarcoma who have limb salvage therapy and reconstruction. Despite being used for larger, more complex defects, the microvascular reconstruction approach provides durable, oncologically sound, functional reconstructions with low morbidity. The greatest effect on function in upper extremity limb salvage with reconstruction appears to result from perioperative complications and potentially the intrinsic nature of the lesions themselves.
FOOTNOTES
Received June 25, 2003; accepted June 16, 2004.
Address correspondence and reprint requests to: David W. Chang MD, Department of Plastic Surgery, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Unit 443, Houston, TX 77030; Fax:713-794-5492; E-mail: dchang{at}mdanderson.org.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
