Oncoplastic Volume Replacement for Breast Cancer: Latissimus Dorsi Flap versus Thoracodorsal Artery Perforator Flap

Background:

Volume replacement oncoplastic breast techniques have become one of the standard lines in the treatment of early breast cancer. They have better cosmetic outcome and patient satisfaction. Latissimus dorsi (LD) flap is one of the most commonly used flaps for these techniques. Although it shows satisfactory surgical outcomes, postoperative shoulder dysfunction is an obvious drawback. The aim of this study was to compare LD flap with thoracodorsal artery perforator (TDAP) flap after breast-conserving surgery regarding surgical outcomes, patient satisfaction, and impact on shoulder function.

Methods:

The study included 42 adult female patients with early breast cancer who were eligible for conservative breast surgery and immediate breast reconstruction. Patients were divided into 2 equal groups: group A where patients underwent immediate reconstruction using LD flap and group B where patients underwent reconstruction using TDAP flap. Follow-up was designed for 12 months for early outcome, patient satisfaction, and shoulder functions.

Results:

The mean age of the included patients in group A and group B was 40.95 ± 5.06 and 40.33± 5.25 years, respectively. There was no significant difference in flap dimensions, postoperative complications, or cosmetic outcome in both groups. However, significantly less shoulder dysfunction was documented in cases of TDAP compared to LD flap at 3, 6, and 12 months postoperatively.

Conclusions:

TDAP flap is as reliable a technique as LD flap regarding the feasibility, postoperative complications, and the cosmetic outcome with significantly better functional outcome of the shoulder.

INTRODUCTION

Multiple meta-analyses and randomized trials have documented the use of conservative surgery and radiation therapy for the treatment of early-stage breast cancer.¹ This conception in breast surgery has developed to attain both minimal surgical intervention and more satisfactory aesthetic results.² It is important for surgeons performing breast surgery to have a basic understanding of which patients are candidates for breast reconstruction and the reconstructive options.³ Excision of more than 20% of the breast volume will increase the risk of worse cosmetic outcome.^4,5 Oncoplastic breast surgery has emerged with the concept of combining tumor excision with clear safety margin followed by breast reconstruction. These procedures include either volume displacement or volume replacement techniques with a clear shift toward immediate reconstruction for a better psychological outcome.^6,7 The aim of breast reconstruction using different oncoplastic techniques is not just creating a mound on the chest wall but achieving symmetry with the contralateral native breast as well.³

Conservative breast surgery with immediate partial reconstruction using the latissimus dorsi (LD) flap has been widely applied as a part of oncoplastic breast surgery. Several studies have documented acceptable surgical outcomes in terms of cosmoses and oncological safety.⁸ It is suitable for patients with small-to-medium-sized breasts and those refusing contralateral surgery. Transposition of the LD flap will not interfere with the subsequent mammogram, because the fatty tissue and the muscle are radiolucent.^9–11 However, functional impairment has been observed in clinical practice following the use of LD flap. This observation has been documented by a lot of studies^12–14 in the form of early postoperative arm and shoulder disability that may interfere with usual daily activities.

The thoracodorsal artery perforator (TDAP) flap is a fasciocutaneous flap that can be an alternative solution. It can offer the theoretical advantage of sparing the LD muscle and thus reducing the donor site morbidity.¹⁵ It is based on a musculocutaneous perforator or perforators from the thoracodorsal artery. The TDAP flap is well suited for head, neck, and extremity defects. A considerable size of the flap can be harvested on a single perforator, with the advantage of both the avoidance of postoperative partial or complete loss of the flap and primary closure of the donor site.¹⁶

The aim of the present study was to compare the TDAP flap with the LD flap in partial breast reconstruction regarding feasibility, cosmoses, postoperative complications, and early musculoskeletal functional outcome.

PATIENTS AND METHODS

The study was conducted at Surgery Departments of the University Main Hospitals in Benha, Menoufia, and Tanta Faculty of Medicine throughout the period from August 2016 to January 2019. Approval to conduct the research was obtained from the institutes’ ethical and research committees. The study included 42 adult female patients diagnosed with early breast cancer (T1, T2) and ductal carcinoma in situ, who were eligible and motivated for conservative breast therapy and immediate breast reconstruction. Exclusion criteria included patients with locally advanced disease, inflammatory breast carcinoma or metastatic disease, those who had a contraindication for radiotherapy or with collagen diseases such as scleroderma, or those with the score of >3 on American Society of Anesthesiologists scale. Written informed consent was obtained from all included patients. Preoperative assessment included full history taking, complete general and local assessment, bilateral mammography, and tissue biopsy. Also, full preoperative laboratory and metastatic work up were done. Included patients were randomized by computer-generated random allocation software, into 2 equal study groups: A and B. Patients in both groups were subjected to conservative breast surgery. Group A included patients for whom immediate reconstruction was performed using LD flap, whereas TDAP flap was used in group B.

For LD flap, preoperative flap design was performed with the patient in sitting position by marking a transverse elliptical skin paddle (Fig. 1). Size of the skin ellipse was adjusted in accordance to the estimated breast defect after conservative breast surgery and to enable primary closure of the donor site. The deep dissection was performed to the thoracodorsal fascia till the separation of the LD muscle from the serratus anterior, paraspinous, and trapezius muscles. The LD muscle was then separated from the humorous after identification of the thoracodorsal artery. A subcutaneous tunnel was created for the transfer of the flap into the breast defect. Finally, closure of the donor site by direct suturing was applied.

Fig. 1.

LD flap. A, Marking of the flap. B, Incision and dissection. C, Full mobilization and tunnel formation. D, Insetting. E, Final aesthetic outcome.

For TDAP flap, preoperative marking of the site of the TDAP was performed using handheld Doppler (Fig. 2). Two anatomical landmarks were determined. The first point would be to the center of the flap and was located 8 cm below the posterior axillary fold and 2 cm behind the lateral border of the LD muscle. This point is corresponding to the site of emergence of the proximal skin perforator from the descending branch of the thoracodorsal artery and its exit off the LD muscle to pass into the SC tissue. The second point was located 3–6 cm below the inferior scapular tip and 1–4 cm medial to the lateral free margin of the LD muscle. This point is corresponding to the site of thoracodorsal artery bifurcation. After the evaluation of volume deficit and location, the TDAP flap was marked in standing position with the arms at sides and hands on waist. It was designed to exceed the lateral edge of the LD muscle and to contain the point of the previously localized artery within its center. As for LD flap, the width of the TDAP flap was designed with the possibility of direct closure of the donor site. Dissection was beveled outward to include the maximum fat, beginning from the anterior side along the supra-fascial plane till pulsation of the perforator was felt and easily observed. When the anterior border of the muscle was reached, a tunnel was created under the lateral breast mound and lateral thoracic wall for passage for the flap in the setting. The vascular pedicle was dissected until enough length was achieved to allow insetting of the flap in the breast defect without tension; then, the donor area was closed directly in 2 layers. Comparison between the 2 groups as regards flap size, operative time, and immediate postoperative complications was performed.

Fig. 2.

TDAP flap. A, Marking of the flap. B, Identification of thoracodorsal artery. C, Full mobilization on the vascular pedicle. D, Insetting. E, Final aesthetic outcome.

Immediate postoperative management included prophylactic anticoagulation in the form of an intravenous infusion of heparin together with prophylactic antibiotic therapy.

Immediate postoperative close monitoring of the viability of the flaps with documentation of any partial or total flap loss was performed.

Postoperative adjuvant therapy was planned for all our patients, and the delivery time was determined to start from 4 to 6 weeks postoperatively to achieve maximum effect. Follow-up for early postoperative complications, patient satisfaction, and range of shoulder movement was done for at least 12 months.

The aesthetic outcome and patient satisfaction were evaluated by patient questionnaire regarding the symmetry of both breasts, the shape of the scar, keloid, and finally the nipple areola complex. This was achieved through a 5-point score (1 = bad, 2 = poor, 3 = fair, 4 = good, and 5 = excellent). The functional outcome of the shoulder was evaluated through Shoulder Pain And Disability Index (SPADI).¹⁷ It was assessed with 8 questions designed to measure the degree of difficulty an individual has with various activities of daily living that require upper-extremity use. To answer the questions, patients were asked to place a mark on a 10-cm visual analog scale for each question. Verbal anchors for the pain dimension are “no pain at all” and “worst pain imaginable,” and those for the functional activities are “no difficulty” and “so difficult it required help.” The scores from both dimensions are averaged to derive a total score. Total disability score is as follows: patient score/80 × 100 = ___%. The original SPADI does not provide specific cutoff points to separate the results into limited, medium, high, or extreme disability. It is considered that the higher the score in each scale, the higher the impairment to the shoulder function. The minimal detectable changes at 90% confidence interval is 13% for the functional disability score. This functional outcome was performed at 3 months postoperatively and repeated at 6 and 12 months. Comparison between the 2 groups was performed.

Statistical analysis was performed using Student’s t test for quantitative parameters that were described using range (minimum and maximum), mean, and SD. Chi-square test was used for qualitative parameters that were described as frequency with percent. SPSS-20 (Statistical Package for Social Sciences version 21) was used. Probability values of <0.05 were considered significant.

RESULTS

The mean age of patients in group A was 40.95±5.06 years whereas it was 40.33±5.25 years in group B (P = 0.699). There was no significant difference between the 2 groups as regards sociodemographic data or comorbidities as shown in Table 1, nor tumor clinical and pathological data as shown in Table 2.The mean operative time was 154.3 ± 11.54 and 155.7 ± 9.26 minutes in group A and group B, respectively (P = 0.661). Table 3 shows no statistically significant difference between the 2 groups regarding postoperative results as hematoma or seroma formation, wound infection or dehiscence. There was no total flap loss in any of the patients in either group. However, partial flap loss has been encountered in 1 patient (4.8%) in group A and in 2 patients (9.5%) in group B (P = 0.99).

Table 1.

Sociodemographic Data and Comorbidities

	Group A (n = 21)		Group B (n = 21)		P
		No.	%	No.	%
Marital status
Single	2	9.5	2	9.5	MCP = 1.000
Married	16	76.2	16	76.2
Divorce	3	14.3	3	14.3
Age, y
Minimum–maximum	31.0–50.0		32.0–51.0		0.699
Mean ± SD	40.95 ± 5.06		40.33 ± 5.25
Median	41.0		40.0
Parity
M	16	76.2	17	81.0	FEP = 1.000
N	5	23.8	4	19.0
Comorbidities
DM	3	14.3	4	19	0.866
IHD	1	4.8	1	4.8	1.000
HTN	2	9.5	3	14.3	0.927

DM, diabetes mellitus; FE, Fisher exact; HTN, hypertension; IHD, ischemic heart disease; MC, Monte Carlo; M, multiparity; N, nulliparity.

Table 2.

Clinical and Pathological Tumor Data

	Group A (n = 21)		Group B (n = 21		P
	No.	%	No.	%
Quadrant					1.000
UOQ	10	47.6	9	42.9
UIQ	4	19.0	5	23.8
LIQ	3	14.4	2	9.5
LOQ	4	19.0	5	23.8
Side					1.000
RT	13	61.9	12	57.1
LT	8	38.1	9	42.9
Size					1.000
T1	7	33.3	6	28.6
T2	14	66.7	15	71.4
Safety margin, mm					0.866
Minimum–maximum	11.0–25.0		12.0–23.0
Mean ± SD	17.52 ± 4.06		17.33 ± 3.17
LN status					0.726
Negative	13	61.9	11	52.4
+1	3	14.3	2	9.5
+2	1	4.8	4	19.0
+3	3	14.3	2	9.5
+4	1	4.8	2	9.5

LIQ, lower inner quadrant; LOQ, lower outer quadrant; UIQ, upper inner quadrant; UOQ, upper outer quadrant; RT, right; LT, left; LN, lymph node.

Table 3.

Postoperative Results

	Group A (n = 21)		Group B (n = 21)		FEP
		No.	%	No.	%
Hematoma	4	19.0	1	4.8	0.343
Seroma	3	14.3	1	4.8	0.606
Infection	2	9.5	1	4.8	1.000
Wound dehiscence	1	4.8	1	4.8	1.000
Partial flap loss	1	4.8	2	9.5	1.000

FE, Fisher exact.

Postoperative hospital stay was calculated from the day of the operation to the day of discharge. It was 7.0 ± 1.22 days for group A and 6.71 ± 0.96 days for group B (P = 0.404). There was no significant difference comparing both groups as regards patient satisfaction for the cosmetic outcome as shown in Table 4 (P = 0.927). Patients showed satisfactory results including “excellent” and “good” outcomes in 80.9% and 76.2% for group A and B, respectively.

Table 4.

Patient Satisfaction

Patient Satisfaction Score	Group A (n = 21)		Group B (n = 21)		MCP
	No.	%	No.	%
Excellent	5	23.8	6	28.6	0.927
Good	12	57.1	10	47.6
Fair	2	9.5	3	14.3
Poor	2	9.5	2	9.5
Bad	0	0.0	0	0.0

MC, Monte Carlo.

Using SPADI, patients in each individual group showed less shoulder function disability and significant improvement with time when compared at 3-, 6-, and 12-month intervals (P < 0.001). Comparing both groups together, patients in groups B had a significantly less shoulder disability compared with group A at the same intervals (P < 0.001) (Tables 5 and 6).

Table 5.

Comparison between the Different Studied Periods According to SPADI

Shoulder Functional Disability	3 mo	6 mo	12 mo	P
Group A (n = 21)
Minimum–maximum	15.0–58.0	9.0–28.0	2.0–16.0	<0.001*
Mean ± SD	28.10 ± 9.78	17.43 ± 5.66	5.62 ± 3.09
Significant between periods	P1 = 0.002*, P2 <0.001*, P3 = 0.001*
Group B (n = 21)
Minimum–maximum	9.0–27.0	4.0–14.0	0.0–5.0	<0.001*
Mean ± SD	17.24 ± 5.36	7.57 ± 3.09	3.05 ± 1.47
Significant between periods	P1 = 0.001*, P2 < 0.001*, P3 = 0.001*

Table 6.

Comparing Both Groups Regarding Shoulder Functional Disability

Shoulder Functional Disability	Group A (n = 21)	Group B (n = 21)	P
3 mo
Minimum–maximum	15.0–58.0	9.0–27.0	<0.001*
Mean ± SD	28.10 ± 9.78	17.24 ± 5.36
6 mo
Minimum–maximum	9.0–28.0	4.0–14.0	<0.001*
Mean ± SD	17.43 ± 5.66	7.57 ± 3.09
12 mo
Minimum–maximum	2.0–16.0	0.0–5.0	<0.001*
Mean ± SD	5.62 ± 3.09	3.05 ± 1.47

DISCUSSION

Oncoplastic breast surgery in the treatment of breast cancer is an intermediate option between conventional breast conservative surgery and mastectomy.⁹ The optimal oncological outcome of conservative breast surgery entails complete excision of malignant tumor with negative resection margins, as involved margins are highly associated with local recurrence.^18,19 However, wide resection can compromise the cosmetic outcome and result in breast deformity or bilateral asymmetry. Oncoplastic breast surgery using volume replacement techniques has the advantage of achieving both wide resection and acceptable cosmetic outcome by partial breast reconstruction.²⁰ As oncoplastic procedure was scheduled for all cases in the current study, a wide safety margin of resection was achieved and there were no involved surgical margins with excised tumors in both groups.

Donor site complications such as hematoma and seroma formation are very common after harvesting flaps for breast reconstruction.²¹ Harvesting of the LD muscle is expected to have more donor site morbidity than flaps with muscle preservation.^22,23 This has been demonstrated by Sowa et al²⁴ who documented that less seroma has developed at the donor site of muscle-sparing LD flap compared with conventional LD flap, with no statistically significant difference. This matches with the results of the current study that showed more incidence of hematoma and seroma formation in group A than group B, although the difference did not achieve statistical significance. In many studies, the incidence of seroma following LD flap ranged from 40% up to 76% in some obese patients.^25–27 However, in the current study, the incidence of seroma in either group was much less. This can be explained by the smaller size of the designed flap required for partial breast reconstruction compared with wider flaps used in those studies for total breast reconstruction. It is quite logical to have the incidence of seroma to be directly related to the size of the dead space left after harvesting the flap.

LD flap is a highly reliable flap with minimal ischemic complications due to the sufficient vascular supply of the thoracodorsal artery.²⁸ Even in patients with diabetes or tobacco use, there is a minimal risk of flap necrosis. Significant flap necrosis is usually secondary to vascular pedicle injury during the operative dissection or pedicle thrombosis from twisting of the flap on its pedicle.²¹ Some studies have reported no flap necrosis at all among the included cases as in the study performed by Lee et al.²⁹ Hokin and Silfverskiold³⁰ reported 7% rate of partial flap necrosis. The same has been documented regarding TDAP flap. Adler et al¹⁵ reported no flap necrosis among the 18 cases of TDAP flap, whereas Angrigiani et al³¹ demonstrated 4.44% total flap loss and 8.88% partial flap loss among the included cases. In the current study, there was no total flap loss in either group; however, partial flap loss occurred in 4.8% and 9.5% in group A and group B, respectively, with no significant difference between the 2 groups.

There is a general agreement that defects following breast-conserving surgery can be managed with primary closure; however, the cosmetic outcome may be unpredictable and frequently patients are unsatisfied.^32–34 Breast-conserving surgery may lead to varying amounts of volume deficit depending on the dimensions of the resected tissue,³¹ with the consequence that approximately 10%–30% of these patients are not satisfied with the aesthetic outcome.³⁵ In the current study, the final cosmetic outcome in both groups was assessed depending on the symmetry, wound scar, and nipple and areola. Satisfied patients for their aesthetic outcome accounted for 80.9% for LD flap and 76.2% for TDAP flap, and none of the patients evaluated their outcome to be bad. These are similar to results reported by other studies using either LD flap^8,29,36 or TDAP flap,^29,31,37 which showed a high percentage of patient satisfaction with either technique. Adler et al¹⁵ explained that the aesthetic result might be somewhat superior with the TDAP flap because of better preservation of the posterior axillary fold. In the current study, there was no significant difference in the patient’s satisfaction comparing the 2 groups.

Although LD flap has demonstrated positive aesthetic outcomes in breast reconstruction with apparent minimal reported complications, postoperative shoulder dysfunction has been overlooked.³⁸ Dejode et al³⁹ in their study demonstrated that LD muscle transfer has sequelae on the ipsilateral shoulder range of movement. However, the exact functional impairment was still a subject of debate. Garusi et al⁴⁰ combined DASH score and objective evaluation of shoulder functions after harvesting LD flap in breast reconstruction and recorded a percentage of recovery. They demonstrated minimal disability in general and up to 80% recovery within 1 year, especially with sports practice.

Blackburn et al³⁸ have documented that breast reconstruction using the LD had an impact on the shoulder function and on some daily life activities, with a significant negative impact not only on the patients themselves but their families as well.³⁸ This significant functional impairment that has been observed in clinical practice was the motivation to explore this area and even to find an alternative to muscle harvesting.¹²

The current study revealed significantly better shoulder functional outcome in cases of TDAP flap compared with LD flap using SPADI. The study also demonstrated interperiodic significant difference within each group separately, during the follow-up at 3, 6, and 12 months, indicating the improvement of the disability in functions of the shoulder in both groups. Despite this individual group improvement, still, the function after TDAP flap was significantly better.

CONCLUSIONS

Our study considers the TDAP flap as reliable a technique as LD flap regarding the feasibility, postoperative complications, cosmetic outcome, and finally early functional outcome, which is significantly better than that of LD flap. There is still a grey area that should be studied further, and we are still in need of algorithm for choosing the appropriate oncoplastic technique depending on the previous parameters.