Since originally described by Hartrampf et al. BREAST

Transcription

1 BREAST Assessment of Zonal Perfusion Using Intraoperative Angiography during Abdominal Flap Breast Reconstruction Albert Losken, M.D. Michael R. Zenn, M.D. Josh A. Hammel, M.D. Mark W. Walsh, M.D. Grant W. Carlson, M.D. Atlanta Ga.; and Durham, N.C. Background: There is an ongoing debate about the reliability of various lower abdominal flaps for breast reconstruction. The authors evaluate in vivo perfusion of these flaps to objectively determine which techniques and which skin island zones had better perfusion. Methods: A prospective study was performed on 77 single-pedicle breast reconstructions [pedicled transverse rectus abdominis muscle (TRAM), n 22; muscle-sparing free TRAM, n 37; deep inferior epigastric perforator (DIEP), n 18]. Perfusion was measured intraoperatively using indocyanine green angiography following flap harvest and before transfer. Flow quantification was performed at 12 standardized data points in each of the four zones of the skin island. Patient risk factors for flap ischemia were assessed, perfusion was quantified, and comparisons were made between the various flaps and between zones. Results: Mean perfusion was significantly higher in the 37 free muscle-sparing flaps (24.9) and the 18 DIEP flaps (21.8) when compared with the 22 pedicled TRAM flaps (19.6) (p 0.001). Zones I and IV had significantly higher and lower perfusion, respectively (28.4 versus 13.9), when compared with the other zones. There was no significant difference in perfusion between zones II and III (20.6 versus 21.6). Differences in flap flow were significant (p 0.001) independent of zonal differences. Conclusions: The authors demonstrated objectively that lower abdominal free flaps based on the inferior epigastric system have better perfusion when compared with pedicled TRAM flaps. There is no appreciable difference in perfusion between zones II and III; however, it is likely related to the perforator location and dominance. Clinical correlation between these absolute perfusion values and flap viability is required. (Plast. Reconstr. Surg. 129: 618e, 2012.) CLINICAL QUESTION/LEVEL OF EVIDENCE: Therapeutic, II. From the Division of Plastic and Reconstructive Surgery, Emory University; and Division of Plastic and Reconstructive Surgery, Duke University. Received for publication July 2, 2011; accepted October 14, Presented at the 54th Annual Scientific Meeting of the Southeastern Society of Plastic and Reconstructive Surgeons, in Naples, Florida, June 4 through 8, Copyright 2012 by the American Society of Plastic Surgeons DOI: /PRS.0b013e b16 Since originally described by Hartrampf et al. in 1982, the transverse lower abdominal tissue flap remains the most common method of autologous tissue breast reconstruction today. 1 In attempts to improve blood supply to this tissue and minimize donor-site morbidity, there have been numerous refinements in flap design and surgical technique. 2,3 Subsequently, knowledge of the anatomy of the lower abdomen, including skin angiosomes, vascular territories, and perforators, has improved. 4 Scheflan and Dinner assigned the original four zones to the unipedicle transverse rectus abdominis muscle (TRAM) flap based on clinical impression of perfusion and reliability. 5,6 Perfusion was felt to decline toward the periphery of the skin island, and the contralateral adjacent zone (zone II) was felt to have better perfusion than the ipsilateral adjacent zone (zone III). Dinner et al., however, later corrected the original description, and with the introduction of the inferiorly based epigastric pedicle (free TRAM flap) and deep inferior epigastric Disclosure: Drs. Losken and Zenn are speakers for LifeCell. Dr. Zenn is a consultant for Novadaq. The other authors have no conflicts of interest to declare in relation to the content of this article. 618e

2 Volume 129, Number 4 Assessment of Zonal Perfusion perforator (DIEP flap), there has been some debate over the accuracy of Hartrampf s original zones of perfusion. 7,8 Some feel that zones II and III should be switched There have been many clinical and cadaveric studies trying to objectivize flow to the various zones of the lower abdominal flaps in either pedicled TRAM flaps, 11 free flaps, 10 or DIEP flaps, 8,12 using a variety of techniques. Fluorescent intraoperative angiography has recently been used to demonstrate tissue perfusion and has the ability of being able to measure and quantify perfusion intraoperatively. 13,14 The purpose of this report was to measure perfusion using this technology within the described zones and compare between the more commonly used lower abdominal flaps. PATIENTS AND METHODS Demographics All women with breast cancer who underwent unilateral (unipedicle) reconstruction using an abdominal flap were included in the series. Institutional review board approval was obtained, and data were collected prospectively. Patient demographics and risk factors for flap ischemia (body mass index, smoking, and diabetes) were queried and the flaps were stratified into (1) pedicled TRAM, (2) muscle-sparing free TRAM, and (3) DIEP flap. The entire skin island was marked corresponding to the original zones (Fig. 1). Zone I overlies the ipsilateral anterior rectus sheath, zone II the contralateral anterior rectus sheath, zone III the ipsilateral and lateral, and zone IV the contralateral and lateral. Angiography Technique Intraoperative angiography was performed once the flap had been completely harvested and just before transfer. Only single pedicle flaps were included, and all four zones were evaluated. A total of 5 mg of indocyanine green was administered in a peripheral intravenous catheter, and angiography was standardized at 1 minute after the saline flush to allow for sufficient tissue perfusion. Absolute flow values were determined using SPY Q (Novadaq Technologies, Inc., Toronto, Ontario, Canada) technology, and 12 data points were obtained in a standardized fashion for each zone (Fig. 1). Comparisons and Statistics To test if there were any differences in flow between pedicle TRAM, muscle sparing free TRAM and DIEP flaps the average of all four zones was calculated per patients and then tested using a Kruskal-Wallis nonparametric test (because of small sample sizes). Subsequent pair-wise testing was used to find which flap type differed from which other specifically. Pair-wise comparisons were performed using Wilcoxon test with a Bonferroni adjustment for multiple testing. Testing for differences between the four zones was performed using analysis of variance. Pair-wise comparisons between the zones were calculated using two-sample t tests with a Bonferroni adjustment. Because of questions about normality and the equality of variances, analysis of variance was also performed on the log-transformed values, but this yielded the same conclusions. Two-way analysis of variance was performed to test whether Fig. 1. A right muscle-sparing TRAM flap with data points positioned per zone. Note good perfusion in all zones. 619e

3 Plastic and Reconstructive Surgery April 2012 there were differences in average flow between the flap types independent of (or controlling for) the differences between the zones. Once again, these were performed using the original and log-transformed values, but results remained the same, so only tests on the original measures are presented. Finally, comparisons were made based on perforator row for the DIEP flaps to assess zonal perfusion. Because of small sample sizes, the Wilcoxon nonparametric test was used to compare medians. All analyses used a 0.05 alpha level of significance and were performed using SAS version 9.2 software. RESULTS There were a total of 77 unipedicle reconstructions using the lower abdominal tissue in this series. Twelve objective perfusion data points were collected for each of the four zones, for a total of 48 per flap. The three groups included DIEP flaps (n 18), muscle sparing free TRAM flaps (n 37), and pedicle TRAM flaps (n 22). The women s age ranged from 30 to 68 years, with an average age of about 50 for all flap groups. There were no significant differences in other ischemic risk factors between the three groups (Table 1). Table 1. Patient Demographics DIEP Flaps Free TRAM Pedicle TRAM n Mean age (SD) 50.7 (7.5) 51.6 (8.3) 48.9 (8.5) Mean BMI (SD) 24.9 (2.8) 27.6 (4.9) 26.7 (4.7) Smoking 0% 3% (n 1) 4% (n 1) Diabetes 0% 6% (n 2) 4% (n 1) BMI, body mass index. Part 1: Testing for Differences between Flap Types The mean perfusion for all four zones per flap indicated that the muscle-sparing free TRAM flaps had the best perfusion (mean SD, ) followed by the DIEP flap (21.2 8) then pedicled TRAM (19.6 7). A Kruskal-Wallis nonparametric test showed that at the 5 percent alpha level, there was evidence of a statistical difference in median flow between the flap groups, with the free TRAM group having a median of 23.5, the DIEP group a median of 18.6, and the pedicled TRAM group a median of 12.6 (Fig. 2). Pair-wise tests show that DIEP and free groups were not statistically different from each other, however, but the pedicle TRAM flap group is different from both of the others (p for free versus pedicle and 0.03 for DIEP versus pedicle). Differences in flap flow were significant independent on zonal differences (analysis of variance not shown; Table 2). Part 2: Testing for Differences between Zones The mean zonal perfusions for the various flaps are shown in Table 2. The average absolute zonal perfusion for all the flaps combined was zone 1 (28.4), zone II (20.6), zone III (21.6), and zone IV (13.9). Analysis of variance demonstrates that zone I had significantly greater perfusion, and zone IV had significantly lower perfusion (p value 0.001; Fig. 3). There was no difference in mean perfusion between zone II and zone III. Because there was some variance between the zones, nonparametric tests were also performed. Kruskal-Wallis test confirmed the differences in medians between groups (p ). Pair-wise Fig. 2. Absolute mean perfusion values for the different flaps. 620e

4 Volume 129, Number 4 Assessment of Zonal Perfusion Table 2. Zonal Perfusion per Flap Type Zone 1 Zone II Zone III Zone IV Mean DIEP (SD) (10.3) (11.1) (8.0) (8.2) Mean free TRAM (SD) (14.4) (11.0) (9.3) (8.2) Mean pedicled TRAM (SD) (19.7) (7.5) (8.5) (5.8) Fig. 3. Distribution of zonal perfusion. comparisons between individual groups showed that zone 1 was different from all others. Zone 4 was significantly different from all others, but there was not a significant difference between zones 2 and 3. Again, two-way analysis of variance showed differences in zonal flow and flap type were independent of each other (analysis not shown; Fig. 4). Part 3: Testing for Zonal Perfusion Depending on Perforator Row There were 18 DIEP flaps. Three had both medial and lateral row perforators. The number of perforators depended on orientation and size, and the decision was made clinically. There were six with only medial row perforators and nine with lateral row perforators. Differences in zonal perfusion are demonstrated in Table 3. Comparisons are made using a Wilcoxon nonparametric test. Although there was a slight improvement in perfusion for zones I and II with medial row perforators and zone III for lateral row perforators, the differences were not significant. DISCUSSION The lower abdominal tissue has become a popular donor site for postmastectomy breast reconstruction. The different zones of perfusion and flap types are often debated, and most surgeons have their preferences. We have been able to quantify intraoperative flow in absolute values to the different zones and make actual perfusion comparisons among the different flap types. There appears to be a perfusion advantage when the deep inferior epigastric system is used and does not appear to be a difference on average between zones II and III for the various flap types. The beneficial use of intraoperative fluorescent angiography (Novadaq) was recently demonstrated in free tissue transfer reconstruction as an easy way of assessing flap perfusion and helping in surgical decision making. 13 Lyophized indocyanine green is injected through a peripheral or central intravenous line. This dye is excited by the laser camera, which results in a fluorescent image on the computer screen. Fluorescence intensity serves as a function of tissue perfusion and real time information is recorded. Although the use of indocyanine green for flap analysis is not new, the ability to quantify flow over an entire zone is new, the result of newer technology and software. Yamaguchi et al. have used indocyanine green imaging in the past to correlate perfusion of the unipedicle TRAM flap and postoperative partial flap necrosis. 11 Although the assessment of fluorescence was more subjective, they did feel that there 621e

5 Plastic and Reconstructive Surgery April 2012 Fig. 4. Intraoperative angiograms of a right unipedicle TRAM flap demonstrating good ipsilateral perfusion and poor zone IV perfusion. Table 3. Medial versus Lateral Row Perforators in DIEP Flaps Medial Row (n 6) Lateral Row (n 9) p Zone 1 median (IQR) 28.8 (9.4) 20.9 (12.0) Zone 2 median (IQR) 19.5 (8.9) 15.8 (14.8) Zone 3 median (IQR) 19.6 (4.3) 19.9 (11.5) Zone 4 median (IQR) 12.8 (3.5) 11.2 (5.6) IQR, interquartile range. was a close relationship between superficial filling of the skin and full flap viability. Holm et al. has also used indocyanine green in DIEP flaps to quantify dynamic flow to the skin island. 8 An intensity number was compared with a well-perfused reference region and a median perfusion index was calculated. They also used time to dye inflow as another dynamic measure for the different zones. Others have measured flap perfusion using cutaneous oxygen levels, computed tomographic angiography, radiography, near infrared reflection spectroscopy, duplex sonography, and laser Doppler flowmetry. 10,12,15 18 We have found the laserassisted indocyanine green fluorescent dye technology (SPY) to be fast, easy to use, and reliable. It provides reproducible real-time information that is both dynamic and quantitative. The deep inferior epigastric system has become a popular alternative to the superior epigastric based pedicle TRAM flap primarily to improve flap perfusion and minimize donor-site morbidity. The deep inferior epigastric system has more direct and larger caliber vessels 19,20 and better flow has been demonstrated within the rectus muscle and the skin when the DIEA is used compared with the superficial inferior epigastric artery flap using ultrasonic flowmetry and thermography. 21,22 Importantly, the inferior epigastric system has been shown clinically to result in fewer complications. 23 Although cadaveric studies help us understand the static lower abdominal anatomy, newer technology has allowed us to quantify flow ex vivo in a dynamic fashion and better understand true flap physiology. The DIEA system represents primary blood supply to the lower abdomen and is clearly dominant to the superior epigastric system, which supplies the pedicle TRAM as a secondary blood supply though a series of choke vessels. This is one of the first studies to compare perfusion in these different flap types and confirms our clinical impression that free TRAM flaps and DIEP flaps have greater perfusion than their pedicle counterpart. We do, however, recognize that although our study confirms improved perfusion in one flap versus another, it has not attempted to correlate this clinically. Our data confirm the findings of previous studies that zone I, benefitting from its underlying 622e

6 Volume 129, Number 4 Assessment of Zonal Perfusion perforators, is the best perfused and most reliable zone in a TRAM flap. We have also confirmed that zone IV is the worst perfused and most unreliable zone in a TRAM flap. This is likely because zone IV requires flow through three choke vessel systems. The relative degrees of perfusion in zones II and III have been more controversial. Holm was the first to perform perfusion studies demonstrating that Hartrampf zones II and III should be switched based on perfusion measurements. 8 Closer evaluation of this study shows that only two of the 15 flaps had perforators based more in the medial row and eight had perforators based more in the lateral row. Recent perforator studies have suggested that the location of the perforator row may have a major impact on zone reliability. 24 This may explain in part the disagreement on perfusion of zone II versus zone III. Because we do not know the dominance of the medial versus the lateral rows in a pedicle flap, those with lateral dominance may favor zone II and those with medial dominance may favor zone III. This concept of dominance and variability in perfusion is supported by the perforasome concept, proposed by Saint-Cyr et al., which describes a distinct vascular territory based on the perfusion of tissue supplied by a particular perforator. 25 Clinical and anatomical variation exist as to the dominance of either the medial or lateral row in terms of flap reliability and design. 12,24,26,27 The ability to measure perfusion intraoperatively will likely improve flap survival in light of these anatomic variations. Zonal perfusion has been quantified in DIEP and free TRAM using oxygen-15 labeled water positron emission tomography. One study of 13 patients showed that there was no difference in perfusion between the flaps and that zone III perfusion was slightly higher than that of zone II. 28 Hallock also looked at free TRAM flaps in 10 patients using laser Doppler flowmetry and demonstrated that flow across the midline was consistently inferior to ipsilateral flow. 10 Ohjimi et al. analyzed the vascular architecture and flow in free muscle sparing TRAM flaps using ex vivo angiography (10 patients) and found that contralateral perfusion was reduced and that the cephalic portion of zone II and all of zone IV had the lowest arterial density. 16 Yamaguchi et al. evaluated blood supply with indocyanine green in pedicle TRAM flaps and presented an individual pattern of perfusion in which zone II was occasionally not as well perfused as zone III. 11 Clinically, it has also been felt that there is an increase in fat necrosis in zone II compared with zone III. 29,30 This was well demonstrated clinically by Kim et al. in 400 unilateral pedicled TRAM flaps, in which fat necrosis in zone II was significantly higher than in zone III. 31 One weakness of this study is the emphasis on arterial perfusion only, ignoring the importance of venous difficulties in determining flap survival. If the clinical risk of fat necrosis in DIEP flaps is twice that of TRAM flaps and perfusion in our study is similar, then perhaps it is the venous component in DIEP flaps that increases the risk and this should be investigated. The egress of indocyanine green dye from flaps detected on SPY angiography may hold this answer in the future. We recognize the limitations of absolute perfusion values in that they are dependent on many factors, including patient age, level of anesthetic, cardiac output, temperature, and so forth. Our study attempted to use sample size and protocol standardization to accurately follow trends in flap and zone perfusion while minimizing any potential confounding influences. Further study correlating flap survival and clinical reliability with perfusion analysis is the next logical step and is underway. Albert Losken, M.D. Emory Division of Plastic Surgery 550 Peachtree Street, Suite Atlanta, Ga alosken@emory.edu REFERENCES 1. Hartrampf CR, Scheflan M, Black PW. Breast reconstruction with a transverse abdominal island flap. Plast Reconstr Surg. 1982;69: Grotting JC, Urist MM, Maddox WA, Vasconez LA. Conventional TRAM flap versus free microsurgical TRAM for immediate breast reconstruction. Plast Reconstr Surg. 1989;83: Allen RJ, Treece P. Deep inferior epigastric perforator flap for breast reconstruction. Ann Plast Surg. 1994;32: Rosen WM, Ashton MW, Taylor GI. Reviewing the vascular supply of the anterior abdominal wall: Redefining anatomy for increasingly refined surgery. Clin Anat. 2008;21: Scheflan M, Dinner MI. The transverse abdominal island flap: Part 1. Indication, contraindications, results and complications. Ann Plast Surg. 1983;10: Scheflan M, Dinner MI. The transverse abdominal island flap: Part II. Surgical technique. Ann Plast Surg. 1983;10: Dinner MI, Downden RV, Scheflan M. Refinements in the use of the transverse abdominal island for post mastectomy reconstruction. Ann Plast Surg. 1983;11: Holm C, Mayr M, Hofter E, Ninovic M. Perfusion zones of the DIEP flap revisited: A clinical study. Plast Reconstr Surg. 2006; 117: Blondeel PN, Arnstein M, Verstraete K, et al. Venous congestion and blood flow in free transverse rectus abdominis and deep inferior epigastric perforator flaps. Plast Reconstr Surg. 2000;106: Hallock GG. Physiological studies using laser Doppler flowmetry to compare blood flow to the zones of the free TRAM flap. Ann Plast Surg. 2001;47: e

7 Plastic and Reconstructive Surgery April Yamaguchi S, De Lorenzi F, Petit JY, et al. The perfusion map of the unipedicle TRAM flap to reduce post-operative partial necrosis. Ann Plast Surg. 2004;53: Rahmanian-Schwarz A, Rothenberger J, Hirt B, Luz O, et al. A combined anatomical and clinical study for quantitative analysis of the microcirculation in the classic perfusion zones of the DIEP flap. Plast Reconstr Surg. 2011;127: Pestana IA, Coan B, Erdmann D, Marcus J, Levin S, Zenn MR. Early experience with fluorescent angiography in free tissue transfer reconstruction. Plast Reconstr Surg. 2009;123: Newman MI, Samson MC. The application of laser assisted indocyanine green fluorescent dye angiography in microsurgical breast reconstruction. J Reconstr Microsurg. 2009;25: Wong C, Saint-Cyr M, Mojallal A, et al. Perforasomes of the DIEP flap: Vascular anatomy of the lateral versus medial row perforators and clinical implications. Plast Reconstr Surg. 2010;125: Ohjimi H, Era O, Fujita T, et al. Analyzing the vascular architecture of the free TRAM flap using intra-operative ex vivo angiography. Plast Reconstr Surg. 2005;116: Scheufler O, Exner K, Andresen R. Investigation of TRAM flap oxygenation and perfusion by near-infrared refection spectroscopy and color-coded duplex sonography. Plast Reconstr Surg. 2004;113: Booi DI, Debats IB, Boeckx RRW, et al. A study of perfusion of the distal free-tram flap using laser Doppler flowmetry. J Plast Reconstr Aesthet Surg. 2008;61: Boyd JB, Taylor GI, Corlett R. The vascular territories of the surperior epigastric and the deep inferior epigastric systems. Plast Reconstr Surg. 1984;73: Hendricks DL, Wilkens TG, Witt PD. Blood-flow contributions by the superior and inferior epigastric arterial system in TRAM flaps, based on laser Doppler flowmetry. J Reconstr Microsurg. 1994;10: Harris NR, Webb MS, May JW. Intraoperative physiologic blood flow studies in the TRAM flap. Plast Reconstr Surg. 1992;90: Salmi AM, Tukiainen E, Aski-Seljavaara S. Thermographic mapping of perforators, skin flow in the free transverse rectus abdominis musculocutaneous flap. Ann Plast Surg. 1995;35: Shusterman MA, Kroll SS, Weldon ME. Immediate breast reconstruction: Why the free TRAM over the conventional TRAM flap? Plast Recosntr Surg. 1992;90: Wong C, Saint-Cyr M, Mojallal A, et al. Perforasomes of the DIEP flap: Vascular anatomy of the lateral versus medial row perforators and clinical implications. Plast Reconstr Surg. 2010;125: Saint-Cyr M, Wong C, Schaverien M, et al. The perforasome theory: Vascular anatomy and clinical implications. Plast Reconstr Surg. 2009;124: Wong C, Saint-Cyr M, Arbique. Three and four-dimensional computed tomographic angiography studies of commonly used abdominal flaps in breast reconstruction. Plast Reconstr Surg 2009;124: Munhoz AM, Ishida LH, Sturtz G, et al. Importance of lateral row perforator vessels in deep inferior epigastric perforator flap harvesting. Plast Reconstr Surg. 2004;113: Schrey A, Kinnunen I, Kalliokoski K, et al. Perfusion in free breast reconstruction flap zones assessed with positron emission tomography. Microsurgery 2010;30: Jewell RP, Whitney TM. TRAM fat necrosis in a young surgeons practice: Is it experience, technique or blood flow? Ann Plast Surg. 1999;42: Shestak KC. Breast reconstruction with a pedicled TRAM flap. Clin Plast Surg. 1998;25: Kim EK, Lee TJ, Eom JS. Comparison of fat necrosis between zone II and zone III in pedicled transverse rectus abdominis musculocutaneous flaps: A prospective study of 400 consecutive cases. Ann Plast Surg. 2007;59: Blondeel PN, Arnstein M, Verstraete K, et al. Venous congestion blood flow in free transverse rectus abdominis myocutaneous, deep inferior epigastric perforator flaps. Plast Reconstr Surg. 2000;106: Carramenha Costa MAC, Carriquiry C, Vasconez LO, et al. An anatomic study of the venous drainage of the transverse rectus abdominis musculocutaneous flap. Plast Reconstr Surg. 1987;79: Man LX, Selver JC, Serletti JM. Abdominal wall following free TRAM or DIEP reconstruction: A meta-analysis and clinical review. Plast Reconstr Surg. 2009;124: e