Comparison of Three Methods for Closure of Mastectomy Incisions in Dogs

CLINICAL REVIEW Comparison of Three Methods for Closure of Mastectomy Incisions in Dogs LG Papazoglou, V Tsioli, M Karayannopoulou, I Savvas, G Kazakos and E Kaldrymidou a Faculty of Veterinary Medicine Aristotle University of Thessaloniki a Department of Clinical Sciences and Laboratory of Pathology Aristotle University of Thessaloniki 11 S. Voutyra Street, 54627 Thessaloniki, Greece ABSTRACT Sixty dogs underwent regional or unilateral mastectomies for the management of mammary tumours. These were randomly allocated into three groups; conventional mastectomy closure (with closed suction drainage), skin-only closure (with closed suction drainage) and a conventional closure group without any drain placement. Conventional mastectomy closure was with a combination of absorbable sutures for dead space elimination and skin sutures. On days one, six and 10 after surgery each incision was graded according to a clinical scale. All dogs were re-examined one month after surgery. Duration of surgery was significantly shorter in the group with skin closure alone, compared to the conventional mastectomy closure groups with drainage (P=0.014) or without (P=0.008). No significant differences were found in respect of complications over time among the three closure techniques. Drain presence or absence had no effect on the outcome. [Papazoglou LG et al (2006) Aust Vet Practit 36:156] INTRODUCTION Canine mammary tumours account for 25 to 50% of all tumours in the bitch (Dorn et al 1968, Brodey et al 1983) and they are the second most common tumours after skin tumours (Kitchell 1995). Early ovariohysterectomy significantly reduces the risk of mammary tumour development (Schneider et al 1969, Taylor et al 1976). Several studies have shown that performing an ovariohysterectomy concurrently with or after surgical excision of the mammary tumour has no effect on survival time (Yamagami et al 1996, Morris et al 1998, Philibert et al 2003). In contrast, two recent studies have found that ovariohysterectomy before surgical excision Ph: 030231099426. Fax: 0302310994400 Email: makdvm@vet.auth.gr (Chang et al 2005), or concurrently with or less than two years before surgical excision, improved survival rate (Soremno et al 2000). Surgical excision remains the treatment of choice for most mammary tumours (Rutteman et al 2002). The aim of surgical treatment is complete tumour removal, with clean histologic margins, by the simplest technique and improvement of prognosis and quality of life (Rutteman et al 2002, White 2003). Closure of mastectomy wounds may be the most challenging part of the surgical procedure and can be achieved by using various techniques. In fact dead space elimination and management of tension on the incision line have been considered the most important issues concerning mastectomy closure (Withrow 1975). Many authors favour conventional mastectomy closure by using 156 Aust Vet Practit 36(4) December 2006 Aust Vet Practit 36(4) December 2006 156

absorbable suture patterns to eliminate dead space (Withrow 1975, Fossum 1997, Harvey 1998, Stone 2000, Waldron 2001, White 2003). In contrast, some others prefer a non-conventional approach to eliminate dead space merely by inserting subcutaneous drains rather than using large amounts of buried suture material (Bright & Aberle 1979, Wilson & Hayes 1983). This study in dogs was designed to compare complications and operative times of mastectomy incisions, closed by three different methods. MATERIALS AND METHODS Sixty dogs admitted to the Veterinary Teaching Hospital, between January 2001 and December 2005, for surgical management of mammary tumours were assessed for this study. Criteria for inclusion were that regional or unilateral mastectomies (MacEwen et al 1985, Allen and Mahafey 1989, Fossum 1997, Stone 2000) were performed for management of solitary or multiple mammary tumours. The size, location and number of tumours determined the extent of the resection. Mastectomies of glands 3-5 and of glands 4-5 were the majority of the regional mastectomies performed. Dogs with inflammatory carcinoma, with tumours grossly invading underlying tissues and those with distant metastases, were excluded. Diagnostic evaluation included history and clinical examination. Thoracic radiography (left, right and ventrodorsal views) was also performed to exclude metastatic disease. Complete blood counts and serum biochemistry panels were also performed. Animals for regional or unilateral mastectomies were premedicated with acetylpromazine 1 (0.05mg/kg im) and butorphanol 2 (0.1mg/kg im). Anaesthesia was induced with propofol 3 (2-3mg/kg iv) and maintained with isoflurane 4 in oxygen. Resection margins were determined pre-operatively. An elliptical incision was performed around the involved mammary glands and a margin of at least 2cm of normal tissue was removed with the tumour. Superficial inguinal lymph nodes were also removed en bloc. Haemorrhage was controlled with electrocoagulation and/or ligation. After excision the surgical wound was thoroughly lavaged with warm normal saline solution. Dogs were then randomly allocated into three groups: conventional mastectomy closure (CMC), skin-only closure (OSC) and conventional non-drain closure (CNDC). In the conventional mastectomy groups (CMC and CNDC) closure was accomplished with a combination of absorbable sutures for dead space elimination and skin sutures. Before closure in the CMC and OSC groups, an active closed drain 5 was placed in the wound, exiting in the inguinal area. The drain was secured to the skin at the exit site with a Chinese finger trap suture and with simple interrupted sutures placed in the skin of the lateral abdominal wall. The CMC and CNDC groups underwent closure of the mastectomy incisions by using simple interrupted walking sutures with 2/0 or 3/0 polyglycolic acid 6 for dead space elimination and tension distribution away from the skin incision line (Swaim 1976). This was followed by a continuous 3/0 gluconate suture 7 for Score CLINICAL SCORING SCALE Description 1 No visible reaction 2 Minimal swelling at the rostral or proximal edge of the incision 3 Suture line inflammation at least 1cm thick with pain and redness 4 Seroma or abscess formation (confirmed by cytology) 5 Dehiscence and clinical treatment 6 Evidence of tumour regrowth (confirmed by cytology and/or histopathology) TABLE 1. Scale for the clinical scoring of mastectomy incisions after closure (modified from Freeman et al 1987). subcutaneous closure and staple 8 placement for skin closure. Closure of incisions in dogs of group OSC was accomplished by tension relieving mattress sutures (Swaim 1980), by using 2/0 polyamide 6 sutures 9, which were pre-placed in the skin along the total length of the incision and then were tightened. No walking or subcutaneous sutures were used for dead space elimination and tension distribution away from the skin incision. Staples were then placed inside each mattress sutures for skin closure. The same surgeons performed the surgery in all groups. In groups CMC and OSC the incisions were then covered with a non-adhesive dressing 10, covered with cotton roll 11, and secured in place with a stockinet bandage 12, that was placed around the thorax, abdomen and inguinal region for the protection of the incision and drain device. The reservoir portion of the drain was secured underneath the stockinet bandage. In group CNDC no coverage of the incisions was performed. Drains were removed as soon as fluid collection in the reservoir bag was minimal. On the sixth post-operative day the bandage was replaced and on Day 10 mattress sutures and staples were removed. Excised tissues were histopathologicly examined on sections stained with haematoxylin- eosin and were classified according to the World Health Organisation scheme (Misdorp et al 1999). On Days 1, 6 and 10 after surgery all dogs were reexamined by the same clinician and each incision was graded according to a scale (Freeman et al 1987), modified for the purpose of this study (Table 1). 1 Calmivet, Vetoquinol, France 2 Butomidor, Richter Pharma AG, Austria 3 Propofol, Abbott 4 Aerrane, Baxter 5 Red-O-Pack, Vygon, France 6 Safil, Braun Aesculap, Germany 7 Monosyn, Braun Aesculap, Germany 8 Appose ULC, USS DG, USA 9 Dafilon, Braun Aesculap, Germany 10 Melolin, Smith & Nephew, United Kingdom 11 Velband, Johnson & Johnson 12 Bend-a-rete, Artsana, Italy 158 Aust Vet Practit 36(4) December 2006 Aust Vet Practit 36(4) December 2006 158

SURGICAL DATA IN 60 MASTECTOMIES Parameters CMC OSC CNDC Age (y) 9.18 ± 2.49 8.38 ± 2.29 9.1 ± 2.32 (range, 5 to 13) (range, 5 to 15) (range, 5 to 14) Weight (kg) 17.45 ± 8.90 19.38 ± 8.67 15.265 ± 9.71 (median 16; (median 17.5; (median 10.75; range, 4.5 to 33) range, 4.6 to 39) range, 5.2 to 37) Mastectomy site left 9 15 10 (no of dogs) Mastectomy site right 11 5 10 (no of dogs) Length of incision (cm) 25.5 ± 6.05 25.5 ± 6.68 24.8 ± 7.49 (range, 15 to 40) (range, 13 to 40) (range, 15 to 40) Duration of procedure (min) 29.30 ± 10.12 22.55 ± 7.62 28.7 ± 8.56 (median 27.50; (median 20; (median 26.5; range, 15 to 60) range, 15 to 40) range, 20 to 45) Volume of fluid collected 19.00 ± 16.40 41.50 ± 46.416 in the reservoir bag (ml) (median 15; (median 20; range, 1 to 60) range, 2 to 160) Duration of drainage (d) 1.55 ± 0.51 1.45 ± 0.60 (range, 1 to 2) (range, 1 to 3) Duration of hospitalisation (d) 2.15 ± 0.74 2.15 ± 0.87 1.2 ± 0.41 (range, 1 to 3) (range, 1 to 4) (range, 1 to 2) TABLE 2. Surgical Data in 60 Mastectomies. Groups (n=20 each) were: CMC conventional mastectomy closure; OSC skin-only closure; CNDC conventional non-drain closure. Data are presented as mean ± SD. Analgesia was provided with postoperative administration of pethidine (3-6mg/kg im) and fentanyl (2-4µg/kg iv) or morphine (0.5-1 mg/kg im) and fentanyl (2-4µg/kg iv) for the first 24 hours. Caprofen 13 (4 mg/kg iv) was administered 30 minutes before surgery followed by 2mg/kg orally twice daily for four days. Cefazolin 14 (30mg/kg iv) was also administered pre-operatively. Dogs were observed closely in the immediate postoperative period and discharged from the hospital from the first to fourth post-operative day. All dogs were re-examined for incisional complications one month after surgery. Data recorded during the process included age, weight, site of mastectomy (right vs. left), type of mastectomy (regional 3-5, regional 4-5 and unilateral), length of incision, duration of surgical procedure (from skin incision to skin closure), volume of fluid collected in the reservoir bag, clinical grade of incision, histologic type of the tumour, duration of drainage and duration of hospitalisation. Data are presented as mean ± standard deviation (SD). Kruskal-Wallis statistics were performed to compare differences among groups in age, weight, and length of incision and to detect homogenous distribution of the mastectomy type among the groups. The Mann-Whitney test was performed to compare differences among the groups in duration of surgical 13 Rimadyl, Pfizer 14 Vifazolin, Vianex, Greece procedure, volume of fluid collected in the reservoir bag, duration of drainage and duration of hospitalisation, as well as to determine the effect of mastectomy site (right vs. left) to the volume of fluid collected in the reservoir bag in either group. The Kruskal-Wallis test was also used to analyse the effect of mastectomy type (regional vs. unilateral) to the volume of fluid collected in the reservoir bag in either group. The number of dogs per clinical grade on Days 1, 6 and 10 were analysed using Pearson chi-square test and Friedman test to determine any differences among or within groups, respectively. Values of P<0.05 were considered significant. RESULTS Clinical parameters of dogs of all groups are presented in Table 2. There were no significant differences regarding age (P=0.428) and weight (P=0.286) among groups. Overall, 34 dogs had left site and 26 dogs had right site mastectomies. Thirty-five dogs had regional mastectomy of glands 3-5, 15 of glands 4-5 and 10 had unilateral mastectomies (Table 3). Mastectomy types had a homogeneous distribution among the groups (P=0.08). No differences were found among groups regarding length of incision (P=0.746). Duration of surgical procedure was shorter in OSC compared to CMC (P=0.014) and to CNDC (P=0.008) but no difference was present between groups CMC and CNDC (P=1). No differences were determined for volume of fluid collected in the reservoir bag (P=0.127) and duration of drainage (P=0.527) between groups CMC and OSC. Duration of hospitalisation was not different between groups CMC and OSC (P=0.973), but was shorter for group CNDC, compared 159 Aust Vet Practit 36(4) December 2006 Aust Vet Practit 36(4) December 2006 159

MASTECTOMIES PER GROUP Type of mastectomy Groups CMC OSC CNDC Regional 3-5 13 12 10 Regional 4-5 4 6 5 Unilateral 3 2 5 TABLE 3. Type of mastectomy per group. Groups (n=20 each) were: CMC conventional mastectomy closure; OSC skin-only closure; CNDC conventional non-drain closure. to group CMC (P<0.001) and group OSC (P=0.001) respectively. Site of mastectomy (P=0.062) and mastectomy type (P=0.739) had no effect on the volume of fluid collected in the reservoir bag in either of groups CMC and OSC. Number of animals per clinical grade on Days 1, 6 and 10 after surgery are presented in Table 4. On Day 10 after surgery, seromas had formed along the incision in one dog of group CMC and in two dogs of group CNDC. These were managed conservatively. A dehiscence (5cm length and 2cm width), which developed on Day 10 in the caudal part of the incision in one dog of group OSC, was debrided and closed primarily with similar sutures. No differences were detected in the number of animals per clinical grade within group CNDC between Days 1 and 6 (P= 0.063) and Days 6 and 10 (P= 0.063), but differences were detected between Days 1 and 10 (P=0.008). Histopathologicly, 54 dogs had malignant tumours (51 carcinomas and three sarcomas) and six had benign tumours. All dogs were re-examined one month after surgery and found to be clinically well. All incisions were inspected and no incisional complications were present. DISCUSSION Most surgeons favour regional or unilateral mastectomies because surgical anatomy is simpler with those approaches (White 2003). The type of surgery does not seem to affect survival (MacEwen et al 1985, Allen & Mahafey 1989) but the size of the primary tumour, its location and fixation to surrounding tissue and the total number of tumours may determine the extent of surgical excision (Rutteman et al 2002). Regional mastectomy is performed when large mammary tumours are located in adjacent glands or a tumour is found between two glands (Allen & Mahafey 1989, Fossum 1997). Additionally, it is sometimes easier and less traumatic to remove both the adjacent caudal abdominal and inguinal glands rather than to perform a simple mastectomy to separate them (Birchard 1995, Fossum 1997). Unilateral mastectomy is commonly selected when multiple mammary tumours occur in most or all glands of a chain, as it is faster and easier to perform than multiple mastectomies (Rutteman et al 2002). In this study, no difference in the length of incisions was found because there was an even distribution of mastectomy-type among the groups and most of the animals belonged to medium and large breeds. The shorter durations of the surgical procedures in group OSC, compared to those of groups CMC and CNDC, might be attributed to placement of fewer sutures and less tissue handling for wound closure. Staples were used for skin apposition in all groups in an attempt to further decrease surgical time. As in the present study considerable tension is often required for skin approximation to close large mastectomy wounds. Walking sutures were used, in groups CMC and CNDC, to move the skin from around the wound to appose skin edges. This helped to distribute tension along the incision, to prevent dehiscence and to aid in dead space elimination (Swaim 1976). Interrupted horizontal mattress tension sutures, that were placed away from the skin edges, were used in group OSC as tension relieving sutures (Swaim 1980). Post-operative complications associated with mastectomy may include seroma formation, wound infection, dehiscence, ischaemic necrosis, selfmutilation, blood loss, hind limb oedema and tumour recurrence. (Birchard 1995, Fossum 1997, Harvey 1998). Regional or unilateral mastectomies may cause considerable amount of dead space, especially in the inguinal area. In the absence of proper drainage, haematoma/ seroma formation may be seen, as a result of inflammation and lymphatic injury, in association with wound oozing because of surgical trauma (Swaim & Henderson 1997, Pavletic 1999). This protein-rich fluid may predispose to bacterial colonisation and abscess development (Bright & Probst 1985). Seroma formation CLINICAL GRADE SCORES CMC Group OSC Group CNDC Group Clinical Grade Day 1 Day 6 Day 10 Day 1 Day 6 Day 10 Day 1 Day 6 Day 10 1 16 15 13 11 13 12 20 15 12 2 4 4 5 9 6 6 0 3 6 3 0 1 1 0 1 1 0 2 0 4 0 0 1 0 0 0 0 0 2 5 0 0 0 0 0 1 0 0 0 TABLE 4. Clinical Grades after Mastectomy. Groups (20 each) were: CMC conventional mastectomy closure; OSC skinonly closure; CNDC conventional non-drain closure. Clinical grades modified from Freeman et al 1987. 160 Aust Vet Practit 36(4) December 2006 Aust Vet Practit 36(4) December 2006 160

may be minimised by following Halsted s principles: careful tissue handling, dead space obliteration and meticulous haemostasis (Mason 1993). In a human study, reporting the effect of dead space closure on seroma formation after mastectomy, it was found that closed suction drainage was significantly less compared to the control group; dead space was decreased in the group that had flaps sutured to underlying muscle and fewer patients in this group developed seromas (Coveney et al 1993). In the present study, dead space was minimised by using closed suction drainage units (group OSC), a combination of tacking sutures and closed suction drainage units (group CMC) or tacking sutures alone (group CNDC). Drains may themselves incite fluid formation, depending on the size of the drain and the size of the wound. As much as 50mL of fluid production can be caused over time by the presence of a drain in a large subcutaneous wound (Fox & Golden 1976, Lee et al 1986). In the present study, in order to identify the necessity of drainage in mastectomy incisions and the effect of the drain on fluid formation a group that had the conventional technique without drain placement was included and comparisons among groups CMC, OSC and CNDC were performed. The findings of the present study support the view that when the conventional technique is used for mastectomy closure, and provided that Halsted s principles are followed, placement of a drain is not necessary. Serious incisional complications were seen in very few dogs of all groups of this study. However, the seroma formation that developed, on Day 10 in one dog of group CMC, might be attributed to inadequate drainage, possibly because of premature discontinuation of the drain or its obstruction. In addition, the seroma formation that developed on Day 10 in two dogs of group CNDC might have been caused by inadequate dead space elimination and not the lack of drainage. Moreover, the dehiscence that developed at the caudal end of the incision in one dog from group OSC might be attributed to an increased tension in the inguinal region. Duration of hospitalisation was significantly shorter for dogs of group CNDC compared to the other groups and this might be attributed to the absence of management issues associated with wound dressing and drain in this group compared to groups CMC and OSC. In conclusion, in terms of surgical time, closure of mastectomy incisions with skin apposition alone, combined with a closed drainage device, was superior to conventional closure techniques. Drain placement or absence of drain had no effect on the outcome. REFERENCES ALLEN, SW & MAHAFFEY, EA (1989) Canine mammary neoplasia: prognostic indicators and response to surgical therapy, J Am Anim Hosp Assoc 25:540 BIRCHARD, SJ (1995) Kirk s Current Veterinary Therapy XII, Ed Bonagura, p462, Saunders, BRIGHT, RM & ABERLE, S (1979) Mammary Neoplasia, Comp Cont Educ Pract Vet 1:774 BRIGHT, RM & PROBST, CW (1985) Textbook of Small Animal Surgery, Ed Slatter, p431, Saunders, BRODEY, RS, GOLDSCMIDT, MH & ROSZEL, JR (1983) Canine mammary gland neoplasms, J Am Anim Hosp Assoc 19:61 CHANG, SC, CHANG, CC, CHANG, TJ & WONG, ML (2005) Prognostic factors associated with survival two years after surgery in dogs with malignant mammary tumors: 79 cases (1998-2002), J Am Vet Med Assoc 227:1625 COVENEY, EC, O DWYER, PJ, GERAGHTY, JC & O HIGGINS, NJ (1993) Effect of closing dead space on seroma formation after mastectomy a prospective randomized clinical trial, Eur J Surg Oncol 19:143 DORN, CR, TAYLOR, DN, FYRE, FL & HIBBARD, HH (1968) Survey of animal neoplasms in Alameda and Contra Costa Counties, California. I. Methodology and description of cases, J Nat Cancer Inst 40:295 FOSSUM, TW (1997) Small Animal Surgery, p539, Mosby, St Louis FOX, JW & GOLDEN, GT (1976) The use of drains in subcutaneous surgical procedures, Am J Surg 132:673 FREEMAN, LJ, PETTIT, GD, ROBINETTE, JD, LINCOLN, JD & PERSON, MW (1987) Tissue reaction to suture material in the feline linea alba. A retrospective, prospective, and histologic study, Vet Surg 16:440 HARVEY, HJ (1998) Current Techniques in Small Animal Surgery, Ed Bojrab, 4th edn, p579, Lea & Febiger, KITCHELL, BE (1995) Kirk s Current Veterinary Therapy XII, Ed Bonagura, p1098, Saunders,. LEE, AH, SWAIM, SF & HENDERSON, RA (1986) Surgical Drainage, Comp Cont Educ Pract Vet 8:94 MASON, LK (1993) Surgical Complications and Wound Healing in the Small Animal Practice, Ed Harari, p33, Saunders, MacEWEN, EG, HARVEY, HJ & PATNAIK, AK (1985) Evaluation of effects of levamisole and surgery on canine mammary cancer, J Biol Response Mod 4:418 MISDORP, W, ELSE, R & LIPSCOMB, T (1999) Histological Classification of Mammary Tumors of the Dog and Cat, p11, Armed Forces Institute of Pathology, Washington DC MORRIS, JS, DOBSON, JM, BOSTOCK, DE & O FARRELL, E (1998) Effect of ovariohysterectomy in bitches with mammary neoplasms, Vet Rec 142:656 PAVLETIC, MM (1999) Atlas of Small Animal Reconstructive Surgery, 2nd edn, p41, Saunders, PHILIBERT, JC, SNYDER, PW, GLICKMAN, N, GLICKMAN, LT, KNAPP, DW & WATER, DJ (2003) Influence of host factors on survival in dogs with malignant mammary gland tumors, J Vet Intern Med 17:102 RUTTEMAN, GR, WITHROW, SJ & MacEWEN, EG (2002) Small Animal Clinical Oncology, Eds Withrow & MacEwen, 3rd edn, p455, Saunders, SCHNEIDER, R, DORN, CR & TAYLOR, DN (1969) Factors influencing canine mammary cancer development and postsurgical survival, J Nat Cancer Inst 43:1249 SOREMNO, KU, SHOFER, FS & GOLDSCHMIDT, MH (2000) Effect of spaying and timing of spaying on survival of dogs with mammary carcinoma, J Vet Intern Med 14:266 STONE, EA (2000) Saunders Manual of Small Animal Practice, Eds Birchard & Sherding, 2nd edn, p222, Saunders, SWAIM, SF (1976) A walking suture technique for closure of large skin defects in the dog and cat, J Am Anim Hosp Assoc 12:597 SWAIM, SF (1980) Management of skin tension in dermal surgery, Comp Cont Educ Pract Vet 2:758 SWAIM, SF & HENDERSON, RA (1997) Small Animal Wound Mangement, 2nd edn, p387, Williams & Wilkins, Baltimore TAYLOR, GN, SAHESTARI, L, WILLIAMS, J, MAYS, WA & McFARLAND, S (1976) Mammary neoplasia in a closed beagle colony, Cancer Res 36:2740 WALDRON, DR (2001) Diagnosis and surgical management of mammary neoplasia in dogs and cats, Vet Med 96:943 WHITE, RAS (2003) Textbook of Small Animal Surgery, Ed Slatter, 3rd edn, p339, Saunders, WILSON, GP & HAYES, HM (1983) Current Techniques in Small Animal Surgery, Ed Bojrab, 2nd edn, p414, Lea & Febiger, WITHROW, SJ (1975) Surgical management of canine mammary tumors, Vet Clin Nth Am 5:495 YAMAGAMI, T, KOBAYASHI, T, TAKAHASHI, K & SUGIYAMA, M (1996) Influence of ovariectomy at the time of mastectomy on the prognosis for canine malignant mammary tumours, J Small Anim Pract 37:462 162 Aust Vet Practit 36(4) December 2006 Aust Vet Practit 36(4) December 2006 162