Wound drainages in total hip arthroplasty: to use or not to use? Review of the literature on current practice

Transcription

1 Musculoskelet Surg (2013) 97: DOI /s REVIEW Wound drainages in total hip arthroplasty: to use or not to use? Review of the literature on current practice M. Nanni F. Perna C. Calamelli D. Donati O. Ferrara A. Parlato M. D Arienzo C. Faldini Received: 4 May 2013 / Accepted: 8 May 2013 / Published online: 25 May 2013 Ó Istituto Ortopedico Rizzoli 2013 Abstract Aim of this study is to analyze data reported in literature concerning the efficacy of using wound lowvacuum suction drainages in orthopedic surgery after total hip arthroplasty. We analyzed studies concerning the use of drainages in prosthetic hip replacement surgery, performing our research through Pubmed, Cochrane database and Google Scholar, and selecting the ones evaluating the following parameters: bleeding, the need for blood transfusions, number or reinforcement of post-operative medications, length of hospitalization, functional results, periprosthetic and surgical wound infection, post-operative hematoma. Our review did not show any demonstrated advantage from the use of wound drainages in total hip arthroplasty. Moreover, some studies enlighted a possible complication related to their employment, represented by the greater need for blood transfusions. Despite the absence of a statistically demonstrated positive influence on wound outcome using suction drains after total hip replacement, many orthopedic surgeons still recommend using drainages, just because there is no certified proof of a negative effect. Keywords Hip arthroplasty Wound drain Surgical drainage Low-vacuum suction drainage Review M. Nanni (&) F. Perna C. Calamelli D. Donati O. Ferrara A. Parlato C. Faldini Dipartimento Rizzoli Sicilia, Istituto Ortopedico Rizzoli, University of Bologna, Bagheria, Bologna, Italy matteonanni@libero.it F. Perna O. Ferrara A. Parlato M. D Arienzo Orthopaedic Department, University of Palermo, Palermo, Italy Introduction Drainages are largely used in all kind of surgeries, especially in general and thoracic surgeries where they are commonly employed in order to collect fluids or air from great cavities. The origins of the use of surgical drains were traced back around 400 B.C., when Hippocrates used a wooden tube to drain an operative wound [18, 21]. In 1961, Waugh and Stinchfield first described drainages use in orthopedic surgery. Although their study showed that there was not any statistically significant difference between the rate of postoperative infections in treated with suction drainages rather than without drainages due to the lack of analyzed cases, the chance of reducing infection or other complications rate, such as hematoma or slow internal bleeding, seemed encouraging [22]. The purpose of using drainages is to reduce the incidence of surgical wounds complications, like hematoma, slow internal bleeding, infections, delayed healing, and wound dehiscence, thus improving postoperative recovery. The use of drainages is also related to risks: indeed, they can represent a path for bacterial colonization of the deepest layers of surgical wounds, as well as foreign bodies capable of compromising the natural defenses of the organism and therefore favoring infections [5]. It is also known that this risk results influenced by the length of drain treatment: the longer the drain is kept into the wound, the more probable is a bacterial contamination [12, 16]. As a matter of fact, most authors suggest to left in place drainages approximately for 24 h [20]. During the last 50 years, the structure of wound drainages has been improved, from open drainage systems to continuous low-vacuum suction systems, provided with a collecting bag isolated by one-way valves and filters, which can allow blood autotransfusion. Currently, in orthopaedic

2 102 Musculoskelet Surg (2013) 97: surgery, the use of low-vacuum suction drainages is preferred, due to their lower absorption rate compared to highvacuum suction drainages with same fluid volume in 24 h, because high-flow suction systems can compromise perilesional tissue s condition as a consequence of their higher irritating effect [2]. Many authors reported no real benefits with the use of drainages in total hip replacement, while other authors suggested the use of surgical drainages in orthopedic surgery, as well as in general surgery, even without a scientific evidence of efficacy. Aim of this study is to analyze the data reported in the literature concerning the efficacy of using wound suction drainages after total hip arthroplasty. Materials and methods For this review, we researched studies concerning the use of drainages in orthopedic surgery. We performed our research through PubMed, Cochrane database, and Google Scholar using the following keywords: hip arthroplasty, surgical drainage, wound drainage, complications, blood loss, blood transfusion, post-operative hematoma, wound dressing and infection. Among the several studies found about surgical drainages, we selected only those ones concerning orthopedics and hip surgery, and between them, we chose the ones which evaluated effective parameters for the determination of a real advantage or disadvantage in the use of wound drainages after a total hip arthroplasty. We considered as significant parameters: bleeding, the need for blood transfusions, number or reinforcement of postoperative medications, length of hospitalization, functional results, periprosthetic and surgical wound infection, and postoperative hematoma. In order to help the analysis of data reported in the selected studies, we calculated percentages where possible, and we also changed unit of measurement of hemoglobin value when it was expressed in g/l to g/dl. Results Bleeding Total hip arthroplasty is considered major surgery operation, which requires a careful control of wound bleeding, in order to restrict the number of blood transfusions needed [1]. In order to determine the blood loss were evaluated the postoperative Hb values and/or the reduction in the average values of Hb compared to the preoperative levels, when the value of the blood loss was not directly measured (Table 1). Most of the studies did not report statistically significant difference between drained and undrained surgical wounds. None of the studies reported evidence in favor of the use of drainages; on the contrary, they speculated that a possible disadvantage in the use of drainages may derive from a higher rate of bleedings [3, 21]. Necessity for blood transfusions One of the possible advantages of using drainages is the chance to reinfuse blood lost from the surgical wound. Three of the selected studies reported a statistically significant higher need of blood transfusions in the group of treated with suction systems after total hip arthroplasty [3, 21]. On the contrary, all other considered studies could not highlight any significant difference after total hip replacement, even though they speculated a possible higher need for blood transfusions with the use of drainages due to the higher incidence of bleeding [8, 10, 14, 16, 17]. Data are reported in Table 2. Number of needed dressings Medication is a procedure that requires time and material, and implies associated costs. It is considered necessary to change dressing whenever it appears dirty due to evident wound leakage; therefore, the number of dressings required is an index of the amount of oozing from the surgical incision and vice versa. The reduced or raised number of needed medications is also an indicator of how drainages affect wound bleeding, even if it does not take into consideration the blood lost from drainage itself. The studies we considered evaluating this parameter reported a statistically significant lower incidence of blood leaking from the surgical wound which resulted in a reduction of needed dressings in groups treated with drainages (Table 3). One of these studies analyzed this parameter by means of Fisher s exact test, which distinguish wound leakage for its absence, slight, or considerable presence [16]. In order to compare results, we considered altogether slight and considerable presence as conditions which led to changed dressing. Length of hospitalization This parameter is mostly related to the morbidity and the time of recovery associated with surgery. Furthermore, a prolonged hospitalization implies higher costs. For this review, we considered studies which compared the length of hospital stay after total hip arthroplasty treated with or without drainage. The majority of them showed no statistically relevant differences in favor or against the use of drainages in terms of length of the hospitalization [3, 17, 18, 21]. One study reported a significant longer hospital stay in treated with drainage, either with or without blood reinfusion [10].

3 Musculoskelet Surg (2013) 97: Table 1 Bleeding Postop Hb Blood loss Postop Hb Blood loss significance levels (g/dl) levels (g/dl) Fall in Hb average (g/dl) Fall in Hb average (g/dl) Walmsley et al. [21] Ovadia et al. [17] Kleinert et al. [10] 552 ( NS ± ± 1.58 NS ( ) 12.9 ( ) NS Biggi et al. [3] S Kim et al. [9] 48 ( (g) (g) S Dora et al. [8] ± (ml) Mengal et al. [14] Murphy et al. [15] Widman et al. [23] ± (ml) (ml) (ml) NS (ml) (ml) S (ml) (1,759 1,025) NS 624 (ml) NS Table 2 Necessity for blood transfusions % of Units of blood % of Units of blood significance transfused transfused transfused transfused Walmsley et al. [21] 552 ( % 26.4 % S Biggi et al. [3] 37 8/12 (66.6 %) 14/25 (56 %) S Ovadia et al. [17] 30 9/18 (50 %) 2/12 (16.6 %) NS Kleinert et al. [10] 80 4/40 (10 %) 4/40 (10 %) NS Widman et al. [23] 22 9/10 (90 %) 6/12 (50 %) S Niskanen et al. [16] U 1.8 U NS Dora et al. [8] ± 1.6 U 2.5 ± 1.9 U NS Mengal et al. [14] U 1.32 U NS Table 3 Number of needed dressings % of needed dressing/no. of dressing per patient % of needed dressing/ no. of dressing per patient significance Niskanen et al. [16] 58 9/31 (29.0 %) 21/27 (77.8 %) S Kim et al. [9] 48 ( % 22.9 % S Dora et al. [8] ± 1.9 (number of dressing per patient) 1.6 ± 1.6 (number of dressing per patient) S Functional results Most studies observed surgical wound outcomes only during patient s hospitalization; therefore, they did not evaluate longtime functional outcome. One study examined this parameter using subjective rating scales for general health assessment, such as Harris Hip Score and SF-36. No statistically significant difference was reported between managed with or without drainages at a medium-term and longterm follow-up (respectively, at 6 months and at 3 years) [21].

4 104 Musculoskelet Surg (2013) 97: Periprosthetic and surgical wound infection Infection is unanimously considered the most devastating complication of prosthetic surgery [7] that may require an aggressive treatment [4, 11]. Therefore, it was evaluated as a relevant parameter for this review. All the studies we examined evaluated the risk of infection, classified in superficial or deep and early or late. The reported rate of infection ranges from 0 to 10 % in treated with drainage and from 0 to 8.33 % in those treated without drainage. None of the considered studies reported statistically significant difference between in which surgical drainage was used and those undrained. Data are reported in Table 4. Postoperative hematoma Hematoma represents an inevitable complication in the majority of orthopaedic surgical procedures, as a result of the difficulty of obtaining a complete hemostasis especially when the medullary canal of a long bone is exposed. In addition, hematoma can increase the tension within the surgical wound or the muscular plans, which can lead to a reduced perfusion of adjacent tissues as much as to a bacterial proliferation [12, 13, 19, 23]. Only few studies reported objective analysis of this parameter. The selected studies evaluated postoperative hematoma considering the incidence rate or reporting the different size of the hematoma (measured in cm 3 ) in treated with or without drainages. Only two of these studies showed statistically relevant difference between the two groups of (Table 5). Discussion The role of surgical drainage in total hip arthroplasty is widely discussed in the literature. We reviewed different studies reported in the literature that correlate some of the main parameters involved in the postoperative recovery or the main complications after total hip replacement with the use of low-vacuum suction drainages, comparing groups of treated either with drainage or without. Main limitations of these studies relate to the relative small number of considered and to usually short followups. In fact, most studies evaluate only during their stay in hospital, and as a result, the literature is characterized by a lack of long-term follow-ups which may be insufficient to evaluate the incidence of rare events, such as infection rate. Moreover, a comparison between the studies and all parameters they analyzed could be difficult as a consequence of different methodologies and evaluation system used. A meta-analysis of the literature by Parker et al. [18] selected 18 prospective and randomized studies in order to examine the benefit of using closed surgical drainages after total hip or knee arthroplasties performed in election. They analyzed 3,495, divided in two groups treated, respectively, with suction drains kept from 24 to 48 h after surgery and without drainages. Authors reported no statistically significant difference between the two groups concerning superficial and deep infection incidence, hematomas, and functional results. On the contrary, they demonstrated that a significant disadvantage for treated with suction drains was an higher rate of bleedings, which resulted also in an increased number of needed blood transfusions. It has to be noted that blood transfusions are associated with a higher risk of transmission of infections and possible immune suppression, which led Parker and others to believe that drainages were of more harm than benefit. On the other hand, they also demonstrated that treatment with drainages was associated with a reduction in blood leaking through the wound, which resulted in a significant reduced number of dressing reinforcements. A randomized prospective study by Ovadia et al. [17] recruited 88 who underwent a total hip or knee arthroplastic surgery and proved no difference in the incidence of infections, in the time of hospital stay and in functional results among who were treated with suction systems for 48 h after surgery or only with medication and bandages. This study appears to confirm Parker s observations showing a statistically significant higher decrease in hemoglobin levels after surgery for Table 4 Periprosthetic and surgical wound infection significance No./% of infection No./% of infection Niskanen et al. [16] NS Ovadia et al. [17] NS Kim et al. [9] 48 ( NS Walmsley et al. [21] 552 ( % (superficial) 4.8 % (superficial) NS 0.4 % (deep) 0.7 (deep) Widman et al. [23] 22 1/10 (10 %) 1/12 (8.33 %) NS

5 Musculoskelet Surg (2013) 97: Table 5 Postoperative hematoma Incidence rate of hematoma/ Incidence rate of hematoma/ significance size of hematoma size of hematoma Walmsleyet al. [21] 552 ( (0.3 %) NS Kim et al. [9] 48 ( % 54.2 % S Kleinert et al ± 2.3 (cm 3 ) 5.3 ± 14.8 (cm 3 ) NS [10] Dora et al. [8] ± 3.3 (cm 3 ) 8.0 ± 17.0 (cm 3 ) S drained. It has to be noted that Ovadia underlined that the use of drainages in hip or knee replacement should be debated separately. In fact, his study showed no statistically significant difference in underwent total hip arthroplasty treated with or without drainages, while a statistically significant higher necessity of blood transfusion for who were treated with drainages after a total knee arthroplasty was reported. Concerning only underwent a total knee replacement, he also found the advantage of a significant reduced serous discharge from the wound, which resulted in a lower number of dressings needed. These observations suggest that drainages were not needed after a total hip replacement, while their use after total knee arthroplasty was still recommended. Similar conclusions were reported also by Biggi et al. [3] which examined 37 who were operated of total hip replacement and the same number operated of total knee replacement. Patients were divided into two groups whether they were treated with two drainages for 24 h or without. Concerning postoperative blood loss, it was pointed out that the medium bleeding after total hip prosthesis was of 3.75 g/dl (range g/dl) with drainage and 2.98 g/dl (range g/dl) without drainage, while for total knee, prosthesis was, respectively, of 3.27 g/dl (range g/dl) and 2.72 g/dl (range g/dl). Furthermore, this study showed an increased number of who needed blood transfusions both after total hip and knee arthroplasty (respectively, 66.6 % drained vs % undrained for hip replacement, and 41.6 % drained vs % undrained for knee replacement). Biggi and others also reported two cases of swelling and hematoma of undrained surgical wounds after a total knee arthroplasty. Nevertheless, the two groups for each kind of joint replacement obtained the same results of functional recovery, joint movement, implant stability, and satisfaction at a follow-up of 6 months. Conversely, some authors still recommend the use of drainages. Kim et al. [9] carried out a randomized prospective study on 48 who underwent a total and bilateral hip arthroplasty, and divided them into two groups whether they were treated on the left or on the right side with two drainages. In order to avoid bias deriving from individual differences, such as sex, age, weight, and height, every patient was also his own control subject. All drains were kept in place for 24 h, and all stayed in bed for 36 h postoperatively. Authors pointed out that postoperatively there was no difference in early complications rate or in functional results, measured with the Harris Hip Score and considering the hip range of motion. In this study was reported that deep and large wound hematomas were significantly more frequent in treated without drainages, while drained were significantly less affected from any surgical wound hematomas. As a matter of fact, this complication should not be underestimated because some authors proved a significant correlation between the incidence of large wound hematomas and the development of an unsatisfactory wound. In addition, Kim underlined a statistically significant higher incidence of wound bruises, rashes, and bleedings when they were treated without suction systems. Hence, those complications required a significantly greater number of medications applied more frequently during the hospitalization. On the other hand, it has to be noted that using more medications is still less expensive than using one drainage: the cost for 4 5 dressings and bed sheet changes was about 50 dollars, while the cost of one drain set was about 135 dollars. Anyways, authors suggested the use of suction systems since the reduced number of medications and less bleedings are psychologically more accepted and preferred by. A randomized prospective study by Niskanen et al. [16] analyzed 58 after total hip arthroplasty and 39 after total knee arthroplasty; were divided into two groups whether they were treated with drainage for 24 h postoperatively or without it. Authors found no differences as concerns mean volume of transfused blood between the two groups after hip surgery, while they enlightened a higher but statistically insignificant number of transfused units in drained after total knee replacement. No differences were observed between the group treated with drainage and the group without it

6 106 Musculoskelet Surg (2013) 97: concerning wound healing time, hematocrit values, and functional results. Niskanen underlined the importance of keeping the drain into the wound for a short period of 24 h, as suggested by the literature, in order to prevent infections due to the bacterial retrograde colonization of the conduit [12, 13, 19]. His research showed no difference between drained and undrained as concerns infection rate. Even in this study, the only advantage of using drainage was to reduce the number of reinforced medications during the hospitalization, in particular after total hip replacement. Kleinert et al. [10] conducted a prospective randomized study on 120 undergoing primary total hip replacement through anterior approach. Patients were divided in three groups: the first group received no drainage, the second group received closed suction drainage, and the last group received two drainages which allowed blood reinfusion. All drains were kept for 48 h, and all wounds were treated with a hydrocolloid drape, which was meant to be kept for 14 days after surgery. Authors reported no differences among the three groups of postoperative hemoglobin values, necessity of transfusions and wound infections, while they reported a higher but not statistically relevant incidence of larger hematomas in undrained wounds which anyway did not affect functional results. Furthermore, this study underlined the significant disadvantage of using suction systems for the longer hospital stay and for the significant higher number of changed dressings in treated with drainage rather than without. Kleinert suggested that these results were strictly influenced by the surgical approach used, which reduces bleedings and is less invasive, but they also concluded that the use of drainages, with or without reinfusion system, does not affect the need of blood transfusion. The most thorough study we found was realized by Walmsley et al. [21] and it is a single randomized trial. They recruited 552 who underwent a total hip arthroplasty, with a medium follow-up of 3 years. Patients were divided into two randomized groups, the first treated with drainage while the second one treated without drainage. The evaluation was based on specific parameters: time of hospitalization, loosening of the implant, surgical revision of the wound or implant, superficial or deep infection, bleeding, transfusions necessity, and functional results measured with Harris Hip Score system. The analysis showed no advantages or disadvantages in the use of drainages, except for the needing of postoperative blood transfusions which were significantly higher in the group treated with drainages. Conclusions From our review of the literature about the use of low-vacuum suction wound drainages after total hip arthroplasty, we can assume that there is not any clear evidence of its efficacy in improving the postoperatory recovery; thus, most authors decided to avoid suction systems. Moreover, some studies enlightened a possible complication related to their employment, represented by a greater need for blood transfusions. Nevertheless, the majority of orthopedic surgeons still rather using drainages, preferring to apply teachings from general surgery, as stated by some authors, despite the absence of a demonstrated efficacy of the use of drainages, just because there is no certified proof of a negative effect [6]. Conflict of interest References None. 1. Ballati S, Giacomello A, Bonoli P, Di Fiore M, Pavone S (1991) Autologous blood pre-deposit in cemented hip arthroplasty. Chir Organi Mov 76(3): Benoni G, Fredin H (1997) Low- or high-vacuum drains in hip arthroplasty? A randomized study of 73. Acta Orthop Scand 68(2): Biggi F, D Antimo C, Trevisani S, Di Fabio S (2005) Efficacia dei drenaggi nella chirurgia protesica standard dell anca e del ginocchio. GIOT 31: Bistolfi A, Massazza G, Rosso F, Ventura S, Lagalla F, Crova M (2009) Modular cementless cup for total hip arthroplasty: result at 4-year follow-up. Musculoskeletal Surg 96: Cerise EJ, Pierce WA, Diamond DL (1970) Abdominal drains: their role as a source of infection following splenectomy. Ann Surg 171(5): Chandratreya A, Giannikas K, Livesly P (1998) To drain or not to drain. Literature versus practice. J R Coll Surg Edin 43: D Angelo F, Negri L, Binda T, Zatti G, Cherubino P (2011) The use of a preformed spacer in two-stage revision of infected hip arthroplasties. Musculoskelet Surg 95: Dora C, von Campe A, Mengiardi B, Koch P, Vienne P (2007) Simplified wound care and earlier wound recovery without closed suction drainage in elective total hip arthroplasty. A prospective randomized trial in 100 operations. Arch Orthop Trauma Surg 127: Kim YH, Cho SO, Kim RS (1998) Drainage versus non drainage in simultaneous bilateral total hip arthroplasties. J Arthroplast 13(2): Kleinert K, Werner C, Mamisch-Saupe N, Kalberer F, Dora C (2012) Closed suction drainage with or without re-transfusion of filtered shed blood does not offer advantages in primary noncemented total hip replacement using a direct anterior approach. Arch Othop Trauma Surg 132: Lamo-Espinosa J, Duart Clemente J, Diaz Rada P, Pons- Villanueva J, Valenti-Nin JR (2013) The Burch-Sneider antiprotrusio cage: medium follow-up results. Musculoskelet Surg 97: Lindahal J, Korkala O, Pammo H, Miettinen A (1993) Bacterial contamination and closed suction drainage in open meniscectomy of the knee. Ann Chir Gynaecol 82: Magee C, Rideheaver GT, Golden GT et al (1976) Potentiation of wound infection by surgical drains. Am J Surg 131: Mengal B, Aebi J, Rodriguez A, Lemaire R (2001) A prospective randomized study of wound drainage versus non-drainage in primary total hip or knee arthroplasty. Rev Chir Orthop Reparatrice Appar Mot 87:29 39

7 Musculoskelet Surg (2013) 97: Murphy JP, Scott JE (1993) The effectiveness of suction drainage in total hip arthroplasty. J R Soc Med 86: Niskanen RO, Korkala OL, Haapala J, Kuokkanen HO, Kaukonen JP, Salo SA (2000) Drainage is of no use in primary uncomplicated cemented hip and knee arthroplasty for osteoarthritis a prospective randomized study. J Arthroplast 15(5): Ovadia D, Luger E, Bickels J, Menachem A, Dekel S (1997) Efficacy of closed wound drainage after total joint arthroplasty a prospective randomized study. J Arthroplast 12: Parker MJ, Roberts CP, Hay D (2004) Closed suction drainage for hip and knee arthroplasty: a meta-analysis. J Bone Joint Surg Am 86(6): Raves JJ, Slifkin M, Diamond DL (1984) A bacteriological study comparing closed suction and simple conduit drainage. Am J Surg 148(5): Spinarelli A, Patella V, Petrera M, Abate A, Pesce V, Patella S (2011) Heterotopic ossification after total hip arthroplasty: our experience. Musculoskelet Surg 95(1): Walmsley PJ, Kelly MB, Hill RMF, Brenkel I (2005) A prospective, randomized, controlled trial of the use of drains in total hip arthroplasty. J Bone Joint Surg Br 87(10): Waugh TR, Stinchfield FE (1961) Suction drainage of orthopaedic wounds. J Bone Joint Surg Am 43: Widman J, Jacobsson H, Larsson SA, Isacson J (2002) No effect of drains on the postoperative hematoma volume in hip replacement surgery: a randomized study using scintigraphy. Acta Orthop Scand 73: