Surgical management of sternoclavicular joint infection

Transcription

1 [(812)TD.STARTITEM] European Journal of Cardio-thoracic Surgery 40 (2011) Surgical management of sternoclavicular joint infection Abstract Walid Abu Arab a,d, *, Ibrahim Khadragui d, Vincent Echavé a, Annie Deshaies b, Chantal Sirois c, Marco Sirois a a Thoracic Surgery Division, Sherbrooke University Medical Center, Sherbrooke, QC, Canada b Orthopedic Surgery Division, Sherbrooke University Medical Center, Sherbrooke, QC, Canada c Thoracic Surgery Division, Hull Hospital, Hull, QC, Canada d Cardiothoracic Surgery Department, Alexandria University, Alexandria, Egypt Received 1 September 2010; received in revised form 29 November 2010; accepted 14 December 2010; Available online 1 March 2011 Objective: Sternoclavicular joint (SCJ) infections are rarely encountered and their management is not well standardised. We reviewed our experience with the management of this condition in order to evaluate the role of surgery in the management of the SCJ infection and to provide an algorithm for its treatment. Methods: It is a multicentre study in which we retrospectively reviewed the data files of the patients who were referred to us for surgical management of SCI infection. Results: From March 2003 to June 2009, 14 patients (12 men and two women) were treated surgically for infected SCJ. No patients were found in the paediatric age group. Mean age was 49.8 years with a range between 26 and 77 years. All patients were symptomatic. The prevalent symptom was either anterior chest wall swelling (21%) or pain (29%); while 50% of them presented with both swelling and pain. Associated risk factors were elicited in 12 patients (86%) while it could not be identified in two patients (14%). These risk factors were in the form of drug addiction in three patients, diabetes mellitus (DM) in four, chronic renal failure (CRF) in three patients and two patients had both DM and CRF. Surgical management was performed in all patients in the form of either incision and drainage in two patients (14%); or SCJ curettage in three patients (21%), while resection of the SCJ was done in nine patients (62%). Mean postoperative hospital stay period (PHS) was 8.1 days (range: 5 30 days). All of them are alive and free of symptoms in follow-up. Conclusion: Surgery was found to be curative with good results for those patients with SCJ infection that did not respond to a full course of intravenous antibiotic therapy. Surgical options include incision and drainage, curettage or SCJ resection. The type of surgical procedure depends on the radiological findings, presentation, severity of the infection and intra-operative findings. In our experience, complex muscle flap reconstruction was not necessary following SCJ resection. # 2011 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved. Keywords: Sternoclavicular joint; Infection; Spontaneous infection 1. Introduction * Corresponding author. Address: Service de Chirurgie Thoracique, Faculte de Médicine et des Sciences de la Santé, Université de Sherbrooke, e avenue Nord, Sherbrooke, J1H 5N4, QC, Canada. Tel.: x12371; fax: address: walidabuarab@yahoo.com (W. Abu Arab). Sternoclavicular joint (SCJ) infections are infrequently encountered. There are only few case reports or small series about surgical management of this entity in the literature. Aetiology and presentation are always variable and are not well characterised [1,2]. Various risk factors were found to be associated with this type of infection. These include diminished immunity, the presence of indwelling central venous catheters, intravenous drug abuse and chronic diseases such as diabetes mellitus (DM), chronic renal failure (CRF) and hepatic dysfunction [3 5]. SCJ infections are particularly challenging to manage because of their proximity to major vascular structures and the lack of substantial overlying soft tissues [2]. Management varies and ranged from conservative management with antibiotics to surgical resection of the joint and may extend to reconstruction with muscle flaps [1,2]. In most cases, the infection responds well to conservative treatment with intravenous antibiotics and local drainage. However, some infections are refractory to these treatment measures. These refractory infections tend to be more extensive and usually spread beyond the boundaries of the joint itself and sometimes invading mediastinal structures [5]. Advances in imaging technology have led to a greater appreciation of the extension of the disease such as abscess formation [6,7], osteomyelitis [8], mediastinitis [9] and empyema [10]. These new emerging tools have important consequences for the optimal medical and surgical management of this unique syndrome [11]. We performed this study in order to evaluate the role of surgery in the management of the SCJ infection and to provide an algorithm for its treatment /$ see front matter # 2011 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved. doi: /j.ejcts

2 W. Abu Arab et al. / European Journal of Cardio-thoracic Surgery 40 (2011) Patients and methods All patients who underwent surgical management for SCJ infections at the Cardio-thoracic Surgery Department, Main University Hospital of Alexandria, Egypt, the Medical center of Sherbrooke University, Canada, and the Hull Hospital, Canada, during the period between March 2003 and June 2009 were reviewed. Medical files were studied with analysis of age, sex, presentation, radiological studies, management plans and outcome. Follow-up was conducted through the outpatient clinic and telephone interviews Surgical technique Anatomy The anatomy of the SCJ is different from others. It is a double joint with two synovial cavities separated by an articular disc. The SCJ is enforced by strong anterior and posterior sternoclavicular ligaments. The range of motion of this articulation is limited in every direction. It plays a role in shoulder movement and stability Incision and drainage Incision and drainage involved performing skin incision directly over the affected SCJ and widely opening the joint capsule. This is followed by drainage of the pus and necrotic tissues. The cavity was irrigated and the wound was either packed and allowed to heal by secondary intention or closed over drains Debridement The same procedure was also followed in those patients who underwent incision, drainage and curettage, whereas the necrotic tissues and the involved osteomyelitic bone were then debrided after the opening of the joint capsule SCJ resection For patients treated with resection of the SCJ, a skin incision was made over the manubrium sterni as far cauded as the level of the third costal cartilage in the midline and carried a variable distance into the supra clavicular region, depending on the exposure required. The soft tissue, pectoralis major muscle and sternocleiodmastoid muscle were dissected to expose the SCJ. The clavicle was then divided lateral to the inflammatory mass with a gigli saw or with reciprocal saw. When indicated, a partial division of the manubrium was then performed with a standard sternal saw or Lebski Knife. This manubrial division preserved about 50% of the manubrium intact. In addition, a small portion of the anterior first rib was sometimes resected. The SCJ was then dissected from the underlying structures and resected en bloc. Further debridement of the clavicle and the manubrium could be performed at this point if the margins of the resections appeared to be not healthy. The residual phlegmon was then carefully debrided to healthy tissues and irrigated. The wound was irrigated and then either packed and allowed to heal by secondary intention or closed over drains. Great care must be taken to avoid subclavian vessels injury leading to serious haemorrhage. The drains were in place for few days (4 12 days). 3. Results There were 14 patients (12 men and two women) included in this study. Age ranged between 26 and 77 years with a mean of 49.8 years. No patients were found in the paediatrics age group. All patients, except one (toe infection); had no coexisting infections and had no other joint sepsis. Three patients were drug addicts, three had CRF, four had DM and two patients had both DM and CRF while two patients had no co-existing medical conditions. Presenting signs and symptoms were elicited in all patients and included anterior upper chest wall swelling (21%), pain (29%) or both of them (50%). One patient (7%) presented with a swelling complicated by sinus formation (Fig. 1). This patient had DM and was operated upon with a diagnosis of subcutaneous abscess. Table 1 demonstrates the age, sex, presenting symptoms, past medical history, radiological findings and final procedure performed for each patient. Radiological studies were performed in all patients. These included plain chest X-ray, computed tomography scan (CT scan) and/or magnetic resonance imaging (MRI), and sonogram when indicated. Chest X-ray was done for all patients and it demonstrated a swelling in nine of them, while it was interpreted as normal in the other five patients. CT scan demonstrated either a swelling alone (nine patients) at sternoclavicular joint or a mass associated with bone destruction (five patients) (Fig. 2). Only one patient, who had a sinus, underwent sonogram to study its extension (Fig. 3). MRI was performed in two patients. Pathological examination and wound culture were done for all patients. They excluded the presence of malignancy and confirmed the diagnosis of SCJ infection. Culture of the wound revealed Staphylococcus aureus in 11 patients (79%), and Mycobacterium tuberculosis in one patient (7%), while it was negative in two patients (14%). Although those patients who were referred to us had a chance with conservative medical treatment, the final curative management for all patients was surgical. Surgery included both incision and drainage alone (14%); or incision and drainage with curettage (21%), whereas SCJ resection was performed for the other nine patients (64%) including the patient who presented with a sinus and another patient who did not respond to the primary management with drainage Fig. 1. Opening of the fistula that extends to the right sternoclavicular joint.

3 632 W. Abu Arab et al. / European Journal of Cardio-thoracic Surgery 40 (2011) Table 1. Demographic table. Patient Sex Age Presentation Antecedent Culture Type of surgical management Radiological finding X-ray/CT/MRI 1 M 34 S + P Addict S. aureus SCJ resection Destructed SCJ 2 M 28 S Addict S. aureus SCJ resection Sever osteomyelitis of the proximal end of the clavicle 3 F 58 S DM M. tuberculosis Incision and drainage Large SCJ effusion 4 M 60 P RF Negative Incision and drainage Moderate SCJ effusion 5 M 26 S + P Free S. aureus SCJ resection Sever osteomyelitic changes in manubrium and clavicle 6 M 77 S + P DM S. aureus SCJ resection Sever erosion of the proximal end of the clavicle 7 M 45 S + P DM S. aureus Curettage Minimal osteomyelitic changes affecting the manubrium 8 M 54 P DM + RF S. aureus Curettage Minimal osteomyelitic changes affection both the proximal end of the clavicle and the manubrium sterni 9 M 47 P DM S. aureus SCJ resection Severely damaged SCJ 10 M 69 S Free S. aureus SCJ resection Deeply eroded manubrium sterni with moderate osteomyelitic changes affecting the proximal end of the clavicle 11 F 70 P RF Negative Curettage Minimal osteomyelitic changes of the clavicle 12 M 45 S + P Addict S. aureus SCJ resection Severely damaged bones of the clavicle 13 M 60 S + P DM + RF S. aureus SCJ resection Sever osteomyelitis 14 M 59 S + P RF S. aureus SCJ resection Sinus formation M: male; F: female, S: swelling, P: pain; S. aureus: Staphylococcal aureus; DM: diabetes mellitus, RF: renal failure. Fig. 2. CT scan in a patient with right sternoclavicular joint infection showing massive pus collection at the right sternoclavicular joint associated with osteomyelitic changes at the medial end of the clavicle and the adjacent manubrium. Fig. 3. Sinogram showing the site and extension of the track from skin to the right sternoclavicular joint. and curettage. Reconstruction with muscle flaps was not needed for any of our patients. All patients received specific antibiotic according to the culture and sensitivity for duration of 2 4 weeks. Postoperative hospital stay ranged between 5 and 30 days with a mean of 8.1 days. No intra- or postoperative complication or mortality was encountered. No recurrence or persistence of infection was encountered in patients who had incision and drainage. Follow-up was complete in all the 14 patients and ranged from 2 to 47 months. All patients are currently alive without symptoms or infection. 4. Discussion The largest series of review of literature included 180 cases which was a review of many small series [11]. SCJ infection can lead to serious complications [6 10,12 14]. Various risk factors were not only reported to be associated with SCJ infections [2,3,15 18], but also reported to occur spontaneously without co-morbidities in healthy patients [19]. In our study, only two patients presented with spontaneous SCJ infection, while the other 12 patients were intravenous drug addicts (three patients) or had DM (four patients) or had CRF (three patients) or had both CRF and DM (two patients). Patients presented with either swelling (21%), pain (29%) or both of them (50%). All patients underwent radiological studies. Chest X-ray demonstrated a mass in 71% of patients; while CT scan/mri demonstrated a mass with its exact localisation and dimensions in all patients. Intrathoracic complications such as mediastinitis or intrathoracic abscess, as some authors mentioned in the literature, were not encountered in this study [11]. The use of radiological tools, especially CTscan or MRI in the diagnosis of SCJ infections was essential as they provided the exact site of the infection, extent of bone destruction and exclusion of intra thoracic extension. We agree with other authors who documented

4 W. Abu Arab et al. / European Journal of Cardio-thoracic Surgery 40 (2011) that plain radiographs are of little importance in the diagnosis of the SCJ infections [12,20]. Culture of the wound revealed S. aureus in 11 patients (79%) in our series which goes with the literature, as it was mentioned that S. aureus causes about half of all sternoclavicular infections. This was explained by the fact that these bacteria have many surface proteins which bind connective tissues [11]. On the other hand, we had only one patient with M. tuberculosis in the culture,(7%) while the culture was negative in two patients (14%). The analysis of retrospective data of clinical presentation and radiological findings helped us to create classification model for SCJI (Table 2) in which, SCJI was classified into five grades depending on symptoms, physical presentation, radiological findings and the presence or absence of mediastinitis. This could help in the future to clear out the keys of diagnosis and management of SCJI. Conservative management with antibiotics in those debilitated patients with SCJ infections is the preferred method by many authors, but usually it results in unfavourable outcomes [12,21]. All of the 14 patients were referred to our department after failure of conservative treatment with antibiotic therapy for several weeks (ranged from 3 to 4 weeks). We considered surgery as the curative management following failure of antibiotic therapy trial, especially when either an abscess, evidence of osteomyelitis, presence of Table 2. SCJI classification. Classification of SCJI Grade Grade I Grade II Grade III Grade IV Grade V Findings Hotness, redness over SCJ Intact skin overlying Mild pain No systemic signs of infection Minimal swelling on X-ray or CT scans No signs of osteomyelitis (intact clavicle, sternum and first rib) Minimal or no effusion at SCJ Moderate to large swelling of SCJ Hotness or redness Pain Systemic signs of inflammation Moderate swelling on X-ray or CT scan with moderate to large effusion No radiological signs of osteomyelitis Any of the criteria mentioned above plus minimal radiological signs of osteomyelitis Any of the criteria mentioned with Grade I or II plus any of the following: (1) Sever osteomyelitic changes (2) Sinus formation (3) Persistence or recurrence of infection Any of the previous grades (I IV) plus evidence of mediastinitis Fig. 4. The algorithm for management of the sternoclavicular joint infection.

5 634 W. Abu Arab et al. / European Journal of Cardio-thoracic Surgery 40 (2011) fistula or sinus or recurrence or persistence of infection. Surgical management included incision and drainage alone, incision drainage and curettage or SCJ resection. The type of the surgical management depended on various factors including the initial presentation of the patient, the patient s general condition, radiological findings and recurrence after initial management. We preferred to start with the least invasive surgery in those patients with poor general conditions, as most of them present with advanced stages of CRFor DM or they were drug addicts. Fig. 4 demonstrates the algorithm we propose in the management of SCJ infection. After resection of SCJ, the wound was closed primarily in most of the cases without encountering any problem and with good cosmetic results. We did not need reconstructive manoeuvre in those patients who underwent SCJ resection. There was no restriction in shoulder movements at follow-up. All patients are doing well without symptoms or recurrence of infection. 5. Conclusion In conclusion, we found that surgery is indicated in cases of SCJ infections after failure of antibiotic therapy trial. The type of the operation depends on the general condition of the patient, and the presence or absence of osteomyelitis. SCJ resection is indicated when there is recurrence of infection, sinus formation, severe osteomyelitis and when there is no response to the other forms of surgical treatment. The newly proposed classification and algorithm of management for SCJI could be of help for thoracic and orthopaedic surgeons in both diagnosis and management of this rare disease. References [1] Bukhart HM, Deschamps C, Allen MS, Nichols III FC, Miller DL, Pairolero PC. Surgical management of sternoclavicular joint infections. J Thorac Cardiovasc 2003;125(4): [(8192)TD.STARTITEM] Editorial comment [2] Song HK, Guy TS, Kaiser LR, Shrager JB. Current presentation and optimal surgical management of sternoclavicular joint infections. Ann Thorac Surg 2002;73: [3] Carlos GN, Kesler KA, Coleman JJ, Broderick L, Turrentine MW, Brown JW. Aggressive surgical management of sternoclavicular joint infections. J Thorac Cardiovasc Surg 1997;113: [4] Yasuda T, Amura K, Fugiwara M. Tuberculous arthritis of the sternocalvicular joint. Phys Ther Rev 1961;41: [5] Haddad M, Maziak DE, Shamji FM. Spontaneous sternoclavicular joint infections. Ann Thorac Surg 2002;74: [6] Linthoudt DV, Velan F, Ott H. Abscess formation in sternocalvicular septic arthritis. J Rheumatol 1989;16: [7] Wohlgethan JR, Newberg AH, Reed JI. The risk of abscess from sternoclavicular septic arthritis. J Rheumatol 1988;15: [8] Tecce PM, Fishman EK. Spiral CT with multiplanar reconstruction in the diagnosis of sternoclavicular osteomyelitis. Skeletal Radiol 1995;24: [9] Pollack MS. Staphylococcal mediastinitis due to sternoclavicular pyarthrosis: CT appearance. J Comput Assist Tomogr 1990;14: [10] Chen WS, Wan YL, Lui CC, Lee TY, Wang KC. Extrapleural abscess secondary to infection of the sternocalvicular joint. J Bone Joint Surg Am 1993;75: [11] Ross JJ, Shamsuddin H. Sternoclavicular septic arthritis: review of 180 cases. Medicine (Baltimore) 2004;83(3): [12] Mozen PH, Zell SC. Sternocalvicular bacterial arthritis. West J Med 1998;148: [13] Asins DS, Dhaliwal GS. Bilateral sternoclavicular joint septic arthritis presenting as cutaneous abscesses. Clin Infect Dis 1994;19: [14] Koroscil TM, Valen PA. Sternoclavicular septic arthritis due to hemophilis influenza. South Med J 1990;83: [15] Buescher TM, Moritz DM, Killyon GW. Resection of the chest wall and central veins for invasive cutaneous aspergillus infection in an immunocompromised patient. Chest 1994;105: [16] Renoult E, Lataste A, Jonon B, Testevuide, Kessler M. Sternoclavicular joint infection in haemodialysis patients. Nephron 1990;56: [17] Covelli M, Lapadula G, Pipitone N, Numo R, Pipitone V. Isolated sternoclavicular joint arthritis in heroin addicts and/or HIV positive patients: three cases. Clin Rheumatol 1993;12: [18] Yood YA, Goldenberg DL. Sternoclavicular joint arthritis. Arthritis Rheum 1980;23: [19] Pairolero PC, Arnold PG, Harris JB. Long-term results of pectoralis major transposition for infected sternotomy wounds. Ann Surg 1991;213: [20] Yüksel C, Kayi Cangir A, Kavukçu S. Spontaneous sternoclavicular joint infection. Tuberk Toraks 2005;53(4): [21] Muir SK, Kinsella PL, Trebilcock RG, Blackstone IW. Infectious arthritis of the sternoclavicular joint. Can Med Assoc J 1985;132: [(812)TD.ENDITEM] Surgical management of sternoclavicular joint infection Keywords: Sternoclavicular joint; Infection; Surgery The recently published article by Arab and colleagues (Ref. [1], in this issue) permits us to discuss sternoclavicular joint infection (SCJI), which is an uncommon and not well-known local infectious disease. Usually occurring in chronic debilitating conditions, after local traumatism or following head-andneck surgery, it might occur in healthy patients in nearly a quarter of cases [2]. Progression might be locally pejorative with complete joint destruction and mediastinal septic contamination. The particular topography of this joint at the cervicothoracic junction and the complexity of this joint partly explain the heterogeneity of care given by different specialists, with different practices. Misdiagnosis and high failure rate of medical therapy are frequently reported in this pathology, including several clinical stages never well defined up to now. For a long time, unrecognized and confused with other joint pathologies, SCJI was first described by Arlet and Ficat in 1958