Donor-site morbidity of the segmental rectus abdominis muscle flap

Transcription

1 British Journal of Plastic Surgery (1998), 51, The British Association of Plastic Surgeons BRITISH JOURNAL OF ~ PLASTIC SURGERY Donorsite morbidity of the segmental rectus abdominis muscle flap M. Geishauser, R. W. Staudenmaier and E. Biemer Department of Plastic and Reconstructive Surgery, Klinikum rechts der Isar, Technical University Munich, Munich, Germany SUMMARY. The donorsite morbidity of the segmental rectus abdominis muscle flap was evaluated in 20 patients with an average followup time of 47 months. Our criteria were based on static and dynamic functional results including relaxation and hernia of the abdominal wall, aesthetic outcome and patient satisfaction. The dynamic functional tests of the abdominal wall showed good results corresponding to the reported minimal impairment of quality of life. There was one abdominal hernia after wound infection and secondary healing. There was no evidence of abdominal wall instability in any of the other patients. The aesthetic outcome was excellent when a transverse lower abdominal incision, asymmetrically elongated to the donor site, was used and moderate in the case of a paramedian vertical incision. Ninetyfive per cent of the patients were completely satisfied or satisfied with the result at the donor site. In the segmental use of the free rectus abdominis flap a high degree of subjective patient satisfaction reflects the favourable outcome of our examinations. On the other hand there is a clinically significant functional donorsite defect of this flap. As this procedure is still widely used, and as its indication is closely linked to its absolute and relative donorsite defect, comparisons with the alternatives, e.g. the partial latissimus dorsi muscle flap, the extended gracilis muscle flap or the serratus anterior muscle flap will have to be made. Free transfer of the segmental rectus abdominis muscle flap is one of the standard procedures for free flap coverage of mediumsize defects, t,2 It is most often used in the lower extremity, especially in the ankle and foot region. Most of the published studies concerning the donorsite morbidity of the rectus abdominis muscle have been done after pedicled or flee TRAMflap surgery for breast reconstruction. 37 Other authors 2,8,9 have reported on the donorsite morbidity of the rectus abdominis muscle flap but without relating it to the amount of muscle taken or without clear functional data. In this study we evaluated the donorsite morbidity of the segmental rectus abdominis muscle flap. In the cases included in this investigation only the caudal segment of one rectus muscle and less than onethird of the total length of the rectus abdominis muscle had been harvested. Our criteria were focussed on static functional results, including relaxation and hernia of the abdominal wall, the dynamic results on the mobility and strength of the abdominal musculature and the effect this has on quality of life, the aesthetic outcome in relation to abdominal shape, position and appearance of the scar, and the overall patient satisfaction as far as the donorsite was concerned. Patients and methods In a retrospective study, we examined 20 out of the 29 patients who had undergone a free transfer of a segmental rectus abdominis muscle flap in our clinic in the years from 1986 to Time of followup ranged from 12 to 110 months with an average of 47 months. Our patient group included 10 male and 10 female patients aged from 24 to 71 years (mean 41 years). Indications for the operation were: open fracture (5 cases), osteitis (9 cases), chronic ulcer (3 cases) and tumour (3 cases). In all patients the recipient area was the lower extremity. Surgical technique The skin incision was carried out as a unilaterally elongated transverse cut in the lower abdomen or as a longitudinal midline or paramedian incision reaching up to the umbilicus. The anterior sheath of the rectus abdominis muscle was then opened, the caudal segmen't of the rectus abdominis muscle was dissected free, including the deep inferior epigastric vessels, and harvested without muscle sparing or splitting. 4 In contrast to the TRAMflap no defect in the anterior rectus sheath resulted. The anterior fascia of the rectus abdominis muscle was then closed by a running suture of 1 Vicryl. No alloplastic material was used to stabilise the abdominal wall and there was no plication of the contralateral abdominal wall. Examination Different methods were used to evaluate the donorsite morbidity of the free segmental rectus abdominis muscle flap: 1. The followup was taken from the patients' chart and case history, including complications and secondary operations. 603

2 604 British Journal of Plastic Surgery 2. Prior to clinical examination, the patients were asked to answer a questionnaire. This was a set of 13 questions for the patients' selfestimation of the functional results, sensation, pain, impairment of quality of life, aesthetic result and overall personal satisfaction with the donorsite. Grades corresponding to the German school mark system could be added. The grades ranged from 1 to 6, 1 representing 'very good', 6 representing 'bad'. 3. Clinical examination was performed focussing on static functional results, especially relaxation or hernias of the abdominal wall, aesthetic result, sensibility and position, appearance and quality of the scar. 4. Function of the abdominal wall was analysed using the test described by Janda:l~ on a stable surface, the patient lies on his back, puts his hands up behind his ears and spreads the elbows laterally. Knee and hip joints are flexed and the soles fixed to the support. Then the patient is asked to lift the head, shoulders and trunk up from the undersurface in a rolling motion. The test is done first in a sagittal plane and then obliquely to the right and to the left side (curlup). In the oblique or rotational curlups the patients were asked to turn the upper body and bring the elbow to the contralateral knee with both hands touching the respective ears. The test was demonstrated to the patient by the investigator and then carried out by the patient once in every plane. The curlups in the sagittal plane are to reflect the function of the rectus muscles, especially in their upper part. Carried out to the side, the ipsilateral obliquus internus and the contralateral obliquus externus muscle are tested, respectively. Care was taken that the examination was performed starting from the described position and in a rolling motion to minimise the influence of synergistic muscles, mainly the iliopsoas muscle. This is in contrast to other tests used for evaluation of the abdominal wall function. 11 The test allows a grading in 6 categories, from grade 0 to grade 5 (Table 1). Results Followup and complications The period of convalescence after operation ranged from 4 to 27 weeks, averaging 13 weeks. This time mainly corresponded to the time to complete healing at the recipient site of the transplantation. There had been one wound infection after a haematoma at the donorsite of the segmental rectus abdominis muscle flap. After secondary healing of the wound, the patient developed a small hernia in the region of the scar, but refused operation for the hernia. Questionnaire The patients' rating of the functional result was very good and good in 14 cases. The impairment in everyday life was rated as moderate by five patients. One patient was unhappy about a slight weakness of the abdominal muscles during gymnastics (Table 2). Three of the patients had had problems getting out of bed. Most of the patients (n = 16) felt no impairment in daily living. With regard to sensation in the donorsite area, the most common complaint was that of moderate numbness in the scarregion. The rating average of sensation was 2.4 (good to moderate). The average rating of the aesthetic result was 2.5 (Table 2). Points for criticism were the longitudinal paramedian and median incision (4 out of 6), the length or the outcome of the scar and asymmetric shape of the lower abdominal wall. Many of the patients stated that, preoperatively, they had felt that a scar on the back might have been more acceptable than on their abdomen. Postoperatively they were highly convinced that with the scar in the lower abdomen they had the more aesthetically acceptable solution. Sixteen patients were glad that they were able to hide the donorsite area including all 14 patients with a transverse incision. The overall personal satisfaction (Table 3) is a collecting pool of multiple individual criteria. Nineteen out of 20 patients were satisfied with the donor site (average rating of 1.6). Regarding the overall longterm result of the operation, 18 patients were satisfied or highly satisfied. Clinical examination In 14 patients a unilaterally extended transverse incision in the lower abdomen was used. Six patients had a longitudinal paramedian or median approach. The length of the scar varied from less than 15 cm (5 cases), between 15 and 20 cm (9 cases), and between 20 and 25 cm (6 cases). We found no keloids, but two moderate hypertrophic scars. An alteration of the abdominal shape was noted in 10 patients. This was much more obvious in patients with less subcutaneous fat than in more obese patients. Two of these patients showed an adherent scar and six others had a palpable laxity in the area where the muscle segment had been removed. One patient had a Table 1 Grade Grading of the muscle strength of the abdominal wall (Janda) Evaluation of the muscle strength of the anterior abdominal wall Lift the trunk including lumbar spine with hands touching the ears and elbows spread laterally Lift the trunk including lumbar spine with arms stretched forward Lift lower scapular angle from support with arms stretched forward Lift cervical spine while pressing the lumbar spine against support Muscle tension palpable only in maximal expiration or while coughing No muscle contraction palpable

3 Donorsite morbidity of the segmental rectus abdominis muscle flap 605 Table 2 Patient's evaluation of the donorsite (1 = optimum, 6 = poor result, n 20) Grade Function Sensation Aesthetic n n n Average grade Table 3 Patient satisfaction concerning donorsite and operation (1 = optimum, 6 poor result, n = 20) Grade Donorsite Operation n n Average grade bulging relaxation and another a small scarhernia after a wound infection and secondary healing. This patient had refused further therapy and felt no discomfort with the scarhernia. In nine patients a hyposensibility of the skin caudally to the donorsite scar was noticed. One patient complained of dysesthesia in the territory of the iliohypogastric nerve, already in regression. There was no evidence of neuroma. Evaluation o f function The function tests of the anterior abdominal wall after Janda were performed as described above. The results of the three steps of the test were averaged for each patient. Additionally, the results for the single movements are shown (Fig. 1). Fourteen patients were able to reach a grading of 5, five patients a grading of 4. Ten patients had a grading of 5 in all three exercises. In one patient we noted a deficit in the oblique exercise to the donorsite. Discussion The transfer of the segmental rectus abdominis free flap is a routine procedure, yet a rare one comprising less than 3% of the microvascular transplantations in our unit. Recent discussions concerning the donorsite morbidity of the free TRAMflap have focussed upon static and dynamic impairment which the loss of even a part of a segment of the rectus abdominis muscle may have on the abdominal wall. This has led to the development of the DIEPflap. On the other hand, the possibility to harvest the segmental rectus abdominisflap endoscopically has encouraged its further clinical use. 12J3 There are reports about incidences of serious weakness and hernias of the abdominal wall in patients undergoing TRAMflap breast reconstruction of up to 35%. 7 This has led to more sophisticated techniques of abdominal wall closure in TRAMflap surgery?,7 In the cases we present here only a unilateral segment of the rectus abdominis muscle was taken without any resulting defect in the anterior rectus sheath. No allograft reinforcement was used, but the anterior rectus sheath was closed tightly and the free borders of the remaining rectus muscle were sutured to the rectus sheath. 100% 80% 60% 40% 20% 0% overall ipsilateral straight contralateral [] Grade 3 [] Grade 4 9 Grade 5 Figure 1Dynamic function of the abdominal wall (n = 20). The resection of a segment of the rectus abdominis muscle leads to an interruption of the muscle. As Blondeel and coworkers ~4 point out, this should lead to a complete loss of function of one of the rectus abdominis muscles. Additionally, this should also lead to major functional changes in the oblique muscles, as their insertion line also will be displaced. Using computed tomography or magnetic resonance imaging, they showed that in TRAMflap patients the midline was deviated to the operated side and the distance between the midline and the insertion line of the oblique muscles was reduced. This can be explained by the resection of muscle and anterior sheath of the rectus muscle that usually is carried out during TRAMflap surgery. In our cases there was no defect in the anterior rectus sheath. As the anterior rectus sheath had to be sutured after taking the segmental rectus abdominis muscle, only a minor displacement of the insertion line of the oblique abdominal muscles, if any, is to be expected. This explains why the abdominal muscle function is only slightly impaired, because the geometry of the abdominal wall does not undergo major changes. The importance of the anterior rectus sheath for abdominal wall stability had already been emphasised by NahaP 5 some time ago. In our study, alteration of the abdominal shape was observed in half of our patients. Six of 20 patients had a laxity where the muscle segment had been removed, which was palpable but not visible. This may be attributed to the missing volume of one segment of the rectus muscle. There was one case of weakness and one incisional hernia after wound infection and secondary healing.

4 606 British Journal of Plastic Surgery Despite this the patient's evaluation of the donorsite was satisfactory. Most of the laxity recorded on critical examination had not been noted by the patients and had no impact on daily life. As noted before, it seems that the static stability of the abdominal wall is attributable to the fact that no defect in the anterior rectus sheath was created. Comparing the results found in our study to the results of Blondeel et ap 6 there seems to be very little difference to the functional results obtained 12 months after the DIEPflap operation, but a big difference to the results obtained with the TRAMflap operation. This holds true for the function in the straight, as well as in the oblique, planes. As only muscle is taken from the abdominal wall in the segmental rectus abdominis muscle flap operation, this operation has an intermediate position between the free TRAMflap operation, where muscle and anterior rectus sheath are resected, and the DIEPflap operation, where muscle is left in place albeit with impaired perfusion as the main vessels are taken away, and the anterior rectus sheath is reconstructed in much the same way as in the segmental muscle flap operation. This raises the question of whether the good functional donorsite results obtained with the DIEPflap should be attributed to the better repair of the anterior rectus sheath rather than to the intact, innervated rectus abdominis muscle. If this could be confirmed, it may be possible to do the TRAMflap operation in an 'anterior rectus sheathsaving' and muscle sparing way, thus saving the extra hour of operating time needed for the intramuscular dissection of the deep inferior epigastric artery. The disadvantage of this retrospective study is the absence of preoperative functional testing of the abdominal wall which in the majority of these patients was not performed. This precludes a comparison of pre and postoperative functional findings. Nonetheless, the results of the dynamic function of the abdominal wall can be considered as good. Taking into account the agerange of our patients and the fact that the strength of the abdominal musculature diminishes with age, it has to be presumed that, even before the operation, some patients would not have reached the maximum grading in these functional tests. This was shown in the preoperative functional tests done by Lejour and Dome ~ and by Blondeel et al. ~6 The clinical functional curlup performance test of the rectus abdominis muscles after Janda ~~ has been discussed already by ZaunerDungl and coworkers 9 and by Blondeel and coworkers? 4 The limiting aspect of most of the functional tests for the abdominal wall done up to now seems to be their inability to determine maximum muscle power strength and fatigue after repetitive contraction. But since the patient's selfestimation compares favourably, they seem to represent a fair indication of impairment of quality of daily life. None of the patients reported new back pain as a result of functional muscle imbalance of the anterior abdominal wall. The loss of sensation in the lower abdominal wall can be considered minimal as it is usually restricted to the region of the scar. The aesthetic results and patient acceptance of the unilaterally extended transverse incision in the lower abdomen were highly convincing. The advantage is the apparent ability to hide the scar during leisuretime activities. The opposite is true for the median or paramedian incisions, so these should be avoided. From our study it can be concluded that there is a functional donorsite defect on removal of the segmental rectus abdominis flap which, despite ongoing discussions, up to now seems to be acceptable. Now that valid functional data concerning the donorsite defect of the alternative surgical procedures such as the partial latissimus dorsi muscle flap, the extended gracilis muscle flap ~7 or the serratus anterior muscle flap ~ are available, comparisons of the relative donorsite defects may be made ~9,2~ to rationalise the differential indications of these alternative flaps. References 1. Cordeiro PG, Neves RI, Hidalgo DA. The role of free tissue transfer following oncologic resection in the lower extremity. Ann Plast Surg 1994; 33: Reath DB, Taylor JW. The segmental rectus abdominis free flap for ankle and foot reconstruction. Plast Reconstr Surg 1991; 88: Hallock GG. Identical rectus abdominis donorsite morbidity in compromised and healthy patients. J Reconstr Microsurg 1994; 10: Hartrampf CR Jr. Abdominal wall competence in transverse abdominal island flap operations. Ann Plast Surg 1984; 12: 139~,6. 5. Kroll SS, Marchi M. Comparison of strategies for preventing abdominalwall weakness after TRAM flap breast reconstruction. Plast Reconstr Surg 1992; 89: Lejour M, Dome M. Abdominal wall function after rectus abdominis transfer. Plast Reconstr Surg 1991; 87: Watterson PA, Bostwick JIII, Hester TR Jr, Bried JT, Taylor GI. TRAM flap anatomy correlated with a 10year clinical experience with 556 patients. Plast Reconstr Surg 1995; 95: Gr6ner R, Feller AM, Biemer E. Donor site morbidity in free rectus abdominis muscle flaps. Handchir Mikrochir Plast Chir 1995; 27:7882 (in German). 9. ZaunerDungl A, Resch KL, Herczeg E, PizaKatzer H. Quantification of functional deficits associated with rectus abdominis muscle flaps. Plast Reconstr Surg 1995; 96: Janda V. Muskelfunktionsdiagnostik, 2nd edn. Berlin: Verlag Volk und Gesundheit, Lacote, M. Evaluation Clinique de la Fonction Musculaire. Paris: Maloine, Friedlander LD, Sundin J. Minimally invasive harvesting of rectus abdominis myofascial flap in the cadaver and porcine models. Plast Reconstr Surg 1996; 97: Miller M J, Robb GL. Endoscopic technique for free flap harvesting. Clin Plast Surg 1995; 22: Blondeel PhN, Boeckx WD, Vanderstraeten GG, et al. The fate of the oblique abdominal muscles after free TRAM flap surgery. Br J Plast Surg 1997; 50: Nahai E Discussion of: Kroll SS, Marchi M. Comparison of strategies for preventing abdominalwall weakness after TRAM flap breast reconstruction. Plast Reconstr Surg 1992; 89: Blondeel PhN, Vanderstraeten GG, Monstrey SJ, et al. The donor site morbidity of fi'ee DIEP flaps and free TRAM flaps for breast reconstruction. Br J Plast Surg 1997; 50: 32~ Holle J, Worseg A, Kuzbari R, Wtiringer E, Alt A. The extended gracilis muscle flap for reconstruction of the lower leg. Br J Plast Surg 1995; 48: Gordon L, Levinsohn DG, Finkemeier C, Angeles A, Deutch H. The serratus anterior freemuscle transplant for reconstruction of the injured hand: an analysis of the donor and recipient sites. Plast Reconstr Surg 1993; 92:

5 Donorsite morbidity of the segmental rectus abdominis muscle flap AskoSeljavaara S, Lahteenmaki T, Waris T, Sundell B. Comparison of latissimus dorsi and rectus abdominis free flaps. Br J Plast Surg 1987; 40: Salmi A, Tuominen R, Tukiainen E, AskoSeljavaara, S. Morbidity of donor and recipient sites after free flap surgery. A prospective study. Scand J Plast Reconstr Surg Hand Surg 1995; 29: The Authors Max Geishauser MD Rainer W. Staudenmaier MD Edgar Biemer MD Department of Plastic and Reconstructive Surgery, Klinikum rechts der Isar, Technical University, Munich, Germany. Correspondence to Dr M. Geishauser. Paper received 17 June Accepted 4 August 1998, after revision.