A Comparison of Pedicled and Free TRAM Flaps for Breast Reconstruction in a Single Institution
David L. Larson, M.D. Department of Plastic Surgery Medical College of Wisconsin | July 1998 |
Thank you for the interesting comments offered by Stephen S. Kroll, Michael J. Miller, David C. Chang, Charles Butler, Gregory P. Reece, Geoffrey L. Robb, and Howard N. Langstein, in their group letter to the editor concerning our article about the cost savings of the DIEP flap in breast reconstruction.
We are in full agreement with the overall conclusions of their letter. Since 1992, when we discovered the clear superiority of the perforator flap for breast reconstruction in terms of decreased morbidity and cost savings for the patient, we have not been able to justify the use of a myocutaneous flap in any breast reconstruction. Thus, for ethical reasons, we were only able to underake one arm of our retrospective review examining the financial cost of sacrificing vs. preserving rectus abdominis muscle in women seeking autologous tissue breast reconstruction. Like the surgeons of MD Anderson, who have since converted to the DIEP flap for 70% of their breast reconstructions, we have found that use of a perforator flap in breast reconstructions has resulted in shorter hospital stays and lower costs for the patient. We thank the surgeons of MD Anderson for pointing out that this finding of shorter hospital stay could potentially be misused by insurance companies to limit post-operative stay to a time period that is medically unsafe. Since the publication of our article, we have also found that overly aggressive early discharge may increase the risk of flap loss, and we have since lengthened the post-operative hospital stay for our patients to 4 days. The bottom line, however, as the surgeons of MD Anderson have agreed, is that patients reconstructed with DIEP flaps experience significantly less pain. This decrease in post-operative morbidity has resulted in shorter hospital stays and cost savings. In the future, we hope that increasing numbers of surgeons might be able to take these findings into consideration when planning autologous tissue breast reconstruction for their own patients.
Robert J. Allen, M.D.
Jonathan Kaplan, M.D.
A Comparison of Pedicled and Free TRAM Flaps for Breast Reconstruction in a Single Institution
The article concludes that the pedicle TRAM is better than the free TRAM with regard to costs and clinical advantages. Our comparison of the DIEP flap to the TRAM flap found the DIEP technique of breast reconstruction to cost significantly less, with shorter operating time and length of stay than the TRAM flap.(1) In addition, Steve Kroll's study of pain medication requirement revealed significantly less pain medication in DIEP patients compared to TRAM patients. I agree with Larson, et al that the free TRAM is not significantly better than the pedicle TRAM. However, in my opinion, the DIEP flap is superior to both the pedicle and free TRAM flaps and should be the procedure of choice for microsurgical breast reconstruction.
R. J. Allen, M.D.
(1) Kaplan, JL and Allen, RJ. Cost-Based Comparison between Perforator Flaps and TRAM flaps for Breast Reconstruction. Plast. Reconstr. Surg. 105: 943,2000.
A Comparison of Pedicled and Free TRAM Flaps for Breast Reconstruction in a Single Institution
David L. Larson, M.D., N. John Yousif, M.D., Raj K.Sinha, M.D., Jorge Latoni, M.D. and Thomas G. Korkos, M.D.
Several reports concluded that free tissue transfer of the transverse rectus abdominis muscle (TRAM) flap for breast reconstruction is superior to pedicled transfer of the flap. In an effort to compare the various parameters of both techniques, the authors took advantage of a unique experience at one hospital where one surgeon (D.L.L.) used only the pedicled method and the other (N.J.Y.) used only free tissue transfer. Additionally, the authors compared the findings of the study with the experieiices of other surgeons by surveying active members of the American Association of Plastic Surgeons.
The records of 119 patients who underwent TRAM flap reconstruction between January of 1988 and July of 1997 were interviewed. Of these, 33 patients received free TRAM flaps, and 86 received pedicled TRAM flaps. To provide an adequate number of patients for statistical analysis, only those with unilateral, single-muscle reconstructions were considered (immediate or delayed). This provided 61 patients in the pedicied flap group and 26 in the free tissue group. Parameters examined included length of operation and of hospitalization, amount of pain medication used, amount of blood lost and received, and complications. A small subset of the patients had hospital records available to compare hospital charges; the comparison of pedicled and 12 free TRAM flaps showed a mean difference of $15,637 (p < 0.001) in favor of the pedicied flap.
On the basis of the findings from this study, it seems that the pedicled TRAM flap has significant economic and clinical advantages over the free TRAM flap. There is less need for blood, a shorter operating time and hospital stay, and a need for less pain medication. However, both methods of transfer have indications and contraindications in certain clinical settings. It will always remain the responsibility of the surgeon to evaluate all issues and select a method that is economically responsible and within the abilities of the surgeon, while producing a satisfactory outcome that best serves the patient. The information provided in this report should aid in accomplishing this goal.
(Plast. Reconstr. Surg. 104: 674. 1999.)
The transverse rectus abdominis musculocutaneous (TRAM) flap, as described by Hartrampf et al., (1) is a standard operation for breast reconstruction. The flap may be transferred as either a free or pedicled flap. There have been various reports on the superiority of one method over the other based on morbidity, ease of surgery, outcome, or cost. In today's health-care environment, these issues are becoming increasingly important.
It seems that the potential for considerable bias exists regarding the method of transferring the abdominal tissue used for breast reconstruction. Ideally, a double-blinded study would provide a definitive answer to the question of which method is the superior technique. In the absence of such a study, we took advantage of a unique experience at one academic institution. One surgeon there used only the pedicled technique (D.L.L.), and the other used only free tissue transfer (N.J.Y.). The purpose of the current study was not to prove the superiority of one technique over the other but, rather, to examine parameters common to both, thereby providing insight on methods to reduce cost, increase efficiency, improve outcomes, and enhance patient care. We also compared our findings with the experiences of other surgeons across the country by using a survey.
One area of particular interest to us was pain control. There is little published about pain control in these patients in the perioperative period. In informal conversations with peers and a random survey of members of the American Associition of Plastic Surgeons, we discovered that some physicians rely solely on oral pain medication in the postoperative period, whereas others use a patient-controlled analgesia device (pump) or even an epidural block.
Materials and Methods
The records of all 119 patients who underwent breast reconstruction with a TRAM flap at Froedtert Memorial Lutheran Hospital in Milwaukee, Wis. between January 1, 1988, and June 30, 1997, were reviewed. All patients were operated on by one of two surgeons. Of these, 86 patients received a pedicled flap (unilateral [61 patients], bilateral [15 patients], or double-muscle [10 patients]), and 33 patients had free tissue transfers (26 patients had a unilateral transfer and seven a bilateral one). To obtain a valid number in both the free and the pedicied groups, a subset of patients was selected. This subset consisted of those patients having unilateral, single-muscle, immediate or delayed reconstruction. The parameters examined included the following: (1) length of operation, (2) days of hospitalization-including the day of surgery, (3) estimated blood loss, (4) amount of blood received, (5) postoperative pain (measured by equianalgesic doses of morphine), and (6) complications. A small subset of patients had hospital charge information available; these charges were used to compare the costs of the 17 pedicled and 12 free TRAM flaps performed since 1996.
Postoperative pain was evaluated by converting the total dose of pain medication received by a patient to an "equianalgesic" total dose of intravenous morphine using the method of Foley.(2) Any adverse occurrence that required secondary surgery or admission, including fat necrosis, was judged a complication.
For the most part, the choice of transfer technique rested with the surgeon who first saw the patient. All patients initially seen by the microsurgeon (N.J.Y.) received a free tissue transfer for breast reconstruction. Of the patients referred to the other senior author (D.L.L.), four were referred for free tissue transfer. The reason for these referrals was patient choice (n = 2) and the need to use the entire abdomen to accomplish the reconstruction (n = 2). No patients were referred for free tissue transfer because of a significant history of smoking, obesity, or diabetes (insulin-dependent). Therefore, the groups were as similar as a nonrandomized group of patients could be under the circumstances.
To determine some form of national standards related to breast reconstruction using the TRAM flap, active members of the American Association of Plastic Surgeons were surveyed. The information regarding the practices of the respondents that was elicited in this survey included the following information: frequency of performance of free versus pedicled TRAM flaps, average length of operation, standard order for controlling postoperative pain, average length of patient's hospital stay, and average amount of blood received by the patient in the perioperative period.
Table I summarizes the findings of the study. The means of the parameters were calculated. Group comparisons were performed with p values calculated using one-way analysis of variance for each variable in both the pedicled and the free transfer groups.
There were 61 patients (mean age, 44.57 years) in the pedicied group and 26 in the free tissue transfer group (mean age 47.84 years). Patients in the pedicled group had an operation that lasted 4.77 hours and were hospitalized for 4.7 days, whereas those in the free tissue transfer group required 8.15 hours of surgery and remained in the hospital for 7.65 days. A mean of 308 cc of blood was lost by patients in the pedicled flap group, with a mean of 61 cc of blood replaced; patients in the free tissue transfer group lost 438 cc of blood on average, and 1000 cc of blood were replaced. Both the hours of surgery and days of hospitalization were significantly different (p < 0.001). The usual postoperative pain order for the pedicled flap group was 50 to 75 mg of intramuscular Demerol (meperidine) each 3 to 4 hours for I to 2 days and then, after a peros (regular) diet was reestablished, Percocet (oxycodone and acetaminophen) was given every 4 hours as required for 5 days. In the free tissue transfer group, standard orders included giving morphine for 1 to 2 days by a patient-controlled pump and then, after a per os (regular) diet was reestablished, Percocet was given each 4 hours as required for 10 days. The amount of pain medication required during hospitalizationi was measured in equidoses of morphine (mg); it was 72.42 mg in the pedicled flap group and 121.32 mg in the free flap group (p < 0.0005). The complication rate was similar in both groups; 18 of 61 patients (30 percent) in the pedicle group and 10 of 26 patients (38 percent) ii-i the free group had complications.The results of the survey are shown in Table II. We had a response rate of 49 percent (138 of 280). Although it was a blinded survey and was only meant to provide a "snapshot" of some parameters of the TRAM flap operation that we reviewed in our study, the survey did provide some useftil information. Of the 126 responders who perform breast reconstruction tissue the TRAM flap (12 responders do not perform the TRAM flap operation), 60 percent (75 of 126) always use the pedicle technique, whereas only, 6 percent (7 of 126) only use free tissue tranfer. The majority of respondents (61 percent; 77 of 126) peform one or two TRAM flap procedures per month; a small number of respondents (17 of 126) perform as many as four or more TRAM flap procedures monthly. The average length of the operation was as low as 70 minutes for one responder and over 8 hours for another, but most operations lasted between these two extremes (46 surgeons, 2 to 4 hours; 65 surgeons, 4 to 6 hours; and 13 surgeons, 6 to 8 hours).
The longer times (in the 6 to 8 hour range) were weighted toward those respondents using free tissue transfer. Two-thirds of surgeons did not give any blood (85 of 126) to patients in the operative or perioperative period, whereas 16 percent of respondents (20 of 126) gave patients 1 U of blood and 15 percent (19 of 126) gave patients an average of 2 U. Two surgeons routinely administered 3 U of blood. Many of the surgeons who routinely gave blood noted that it was autologous. When asked for their standard postoperative order for pain medication within the first 2 or 3 days. 50 percent of surgeons (63 of 126) responded that they use a patient-controlled pump, 5 percent (six surgeons) use an epidural anesthetic, 3 percent (four surgeons) used only oral pain medication, and the remainder (42 percent; 53 of 126) use intravenous or intramuscular morphine or Demerol. The majority of the surgeons who perform free tissue transfer use a patient-controlled analgesia pump, whereas only 40 percent of the surgeons performing pedicled flaps use the device. Most responders had patients stay in the hospital for 3 (31 of 126), 4 (36 of 126), or 5 days (44 of 126), including the day of surgery. The free flap patients seemed to require longer hospital stays. The extremes for hospital stays were 2 days (3 surgeons) and 7 days (1 surgeon). The pain medication received upon discharge was split among Vicodin (37 surgeons), Percocet (61 surgeons), and Tylenol 3 or Darvocet (28 surgeons). These medications were needed for pain control for up to 2 weeks in many patients.
Hospital charges were only available from 1996 to the present; we used these charges to compare 17 pedicled and 12 free TRAM flaps. The comparison showed a mean difference of $15,637 (p < 0.001) in favor of the pedicled flap.
Since it was initially described by Hartrampf et al.(1) in the early 1980s as a pedicled flap, the TRAM flap has steadily gained popularity worldwide as the "gold standard" for breast reconstruction. within a short time, others expanded the method of transfer to the free tissue technique.(3-6) By the late 1980s. substantial debate existed regarding the merits of the "conventional" TRAM flap and the "microsurgical" TRAM flap.(3-6) As stated by Schusterman (4) and Grotting et al.,(3) the advantages of free tissue transfer in the immediate setting include the ready availability of the recipient vessels for atastomosis, improved blood supply, with a larger, more predictable vessel (which avoids or prevents fat necrosis and partial skin loss), availability when the upper abdomen is scarred, reduction of the incidence of hernia, and retention of the entire inframammary fold. Its disadvantages include longer operating time, a need for special expertise, the "all or none" phenomenon of microvascular surgery, (5-7) and some technical problems in a patient receiving a delayed reconstruction (e.g., need for vein grafts and occasional need for dissection of internal mammary vessels). The usual patient seen for breast reconstruction is a healthy, young or middle-aged woman with a negative medical history who has a medium build and an adequate abdominal pannus to provide semmetry with the opposite breast; she is a good candidate for either method of transfer.
Resource Cost Comparison
One area of comparison between the two TRAM techniques in breast reconstruction is related to resource cost. The most extensive study of this topic was reported by physicians from the M. D. Anderson Cancer Center Department of Reconstructive and Plastic Surgery.(8-9) They found that autogenous tissue reconstruction with the TRAM flap (primarily free tissue transfer) was at least as cost effective, in terms of time and dollars, as implant-based reconstruction. On the basis of these studies, we may assume something that is counterintuitive: all standard methods of breast reconstruction cost approximately the same when considering resource consumption.
The financial records of the TRAM patients at our institution were only available from 1996 to the present. Recognizing that data exist only for 17 pedicled and 12 free TRAM flaps, a significant difference of over $15,000 still existed, which favored the pedicled procedure. Most of these charges were related to the longer hospital stay and, to a lesser degree, the longer operative time required for free TRAM flaps.
Recently, Edsander-Nord et al.(10) prospectively studied 23 patients with pedicled TRAM flaps and 19 with free TRAM flaps for breast reconstruction. Using both a patient questionnaire and an active isokinetic dynamometer system to study maximal voluntary trunk flexion and extension preoperatively and at 6 and 12 months postoperatively, they determined that the method of transfer had no influence on postoperative abdominal wall strength per se.
Pain Control Comparison
As shown from conversations with peers and the results of our survey (Table II), a wide range of methods for controlling postoperative pain in the TRAM-flap patient seems to exist. As noted in the survey, some surgeons use nothing more than oral pain medication supplemented with parenteral or intravenous drugs, whereas others use a patient-controlled analgesia pump or epidural anesthesia. One would think that there should be more agreement than this, despite the diversity of preferences!
The use of equianalgesic doses of drugs is common when treating cancer pain to prevent undermedication when switching from one route of administration to another.(2) Because this is the standard in clinical oncology and pain clinics, it seemed appropriate to apply this concept to our postoperative TRAM patients to compare analgesic requirements using a variety of analgesics.
Final Selection Is the Surgeon's Preference
The success of either TRAM operation can be improved by appropriate patient selection, patient education, and a noninvasive Doppler vascular study of the abdominal wall to map out the location and integrity of the vessels.(11) As Clark (l2) states in his discussion of the article by Schusterman et al(4)
... most investigators use their best professional judgment to choose possible confounding factors and then fit a logistic regression equation where the outcome variable is successful treatment or not and the predictor variables include both which treatment the patient received and the possible confounding variables. For this model to work, the logistic regression equation must actually correctly model the real world, and all important confounding factors must be included in the model.
In other words, surgeons must use their clinical judgment to provide patients with the best result possible within their surgical capabilities.
Significance of This Study
We have taken advantage of a unique situation at one hospital-one surgeon (D.L.L.) performing pedicle TRAM flaps exclusively and another (N.J.Y.) using only free tissue transfer. There were four referrals for free tissue transfer (two by patient request and two because the entire abdomen was needed); all of these patients were otherwise healthy women. There was no statistically significant difference between the two methods of transfer with regard to complications. There was statistical difference between patient age; patients in the free tissue group were 3 years older. Significant differences also existed in the length of the operation, number of days in the hospital, and the amount of pain medication required after surgery. Almost all patients received a second surgery for nipple reconstruction, mound revision, and/or obtaining symmetry with the opposite breast. Because the breast mound provided by the free TRAM flap has a richer blood supply, it can be sculpted with more confidence and it has a more "finished" look after the initial surgery. There were more mound "revisions" as the second surgery after the pedicle procedure, but the end results of both procedures were judged comparable. Comments have been made in conversation and implied in the literature that the use of the pedicled TRAM flap "sacrifices" the abdominal wall. This view is used as further justification for the use of the free TRAM flap. For those who use a single-muscle TRAM flap associated with minimal fascia harvest (2 to 3 cm) and leave the lateral third of the rectus muscle (which has an intact blood and nerve supply (l3) ) when using pedicied TRAM flaps, no advantage of one method of transfer over the other seems to exist. (10)
It seems that the TRAM flap patient would benefit from less pain medication delivered in a more economical manner (intramuscularly or intravenously versus a patient-controlled pump) given over a shorter time frame (1 or 2 days).
One factor that might have played a role in the reduced analgesic requirements of the patients with pedicled flaps in our review is the surgeon's (D.L.L.) empiric use of 500 mg of intravenous methylprednisolone at the time of flap elevation. Admittedly, little scientific evidence exists for administering this drug,(14-16) but it seems to provide the patient with a decrease in the initial need for parental analgesia by, providing a psychological "boost" and making the transition to oral pain medication more seamless. If patients are properly selected (eliminating patients with tuberculosis, ocular herpes simplex, acute psychosis, or history, of gastrointestinal bleed), the only disadvantage of this single dose of steroid is the possibility of a mild, self-limiting depression, which occasionally occurs 5 to 10 days postoperatively.
Our hospital found these data valuable in establishing resource consumption, minimizing practice variations, and improving surgical efficiency.
It seems that both methods of TRAM flap transfer for breast reconstruction have inherent advantages and disadvantages. Most of the literature to date has suggested that the pedicled and free tissue transfers are equal in relation to outcome, result, and cost. There is little question that there are certain indications for free tissue transfer. These include a significant history of smoking, obesity; diabetes (insulin-dependent); preexisting, unfavorable abdominal scars; or required use of the entire abdomen for reconstruction. Regardless, the majority of women requesting TRAM flap reconstruction of a breast do not fall into these categories and can, in fact, receive a pedicled flap. With this in mind, the surgeon should tailor his or her choice for method of transfer to the specific patient rather than use one method for all. Our findings suggest that all TRAM patients could have perioperative pain in the first or second day controlled by a simple, parental analgesics, without the need for an analgesia pump or epidural analgesia. Although little has been written on the advantages of the pedicled TRAM flap over free TRAM transfer, our experience indicated that with the pedicled TRAM flap, there is less need for blood and a statistically significant difference when comparing the length of the operation and of hospital stay. Each of these parameters would, of course, translate into decreased hospital cost. This cost factor must be considered in today's health-care environment, especially because recent federal legislation requires insurers to offer breast reconstruction to all mastectomy patients but makes no mention of reimbursement being guaranteed to the provider.
Regardless, it must remain the responsibility of the surgeon to evaluate each patient's needs and select a method that is economically responsible, within his or her abilities, and that can produce a satisfactory outcome that best serves the patient. It is hoped that this information aids in that decision.
David L. Larson, M.D.
Department of Plastic Surgery
Medical College of Wisconsin
9200 West TVisconsin Ave.
Milwaukee, Wis. 53226