The Knee 17 (2010) 176–178

Contents lists available at ScienceDirect

The Knee

Short Communication

Mycobacterium fortuitum infection after anterior cruciate ligament reconstructionusing a polylactic acid bioabsorbable screw: Case report

Horng Lii Oh, Darren B. Chen, Bradley G. Seeto, Samuel J. MacDessi ⁎Sydney Knee Specialists, Suite 211, 203-233 New South Head Road, Edgecliff, NSW 2027, Australia

⁎ Corresponding author. Tel.: +61 2 8307 0333; fax:E-mail address: samuelmacdessi@sydneyknee.com.a

0968-0160/$ – see front matter © 2009 Elsevier B.V. Adoi:10.1016/j.knee.2009.08.004

a b s t r a c t

a r t i c l e i n f o

Article history:Received 29 March 2009Received in revised form 6 August 2009Accepted 10 August 2009

Keywords:AbscessAnterior cruciate ligament reconstructionBioabsorbable interference screwInfectionMycobacterium fortuitum

We report a case of pretibial sinus and abscess after anterior cruciate ligament reconstruction using apolylactic acid tricalcium phosphate bioabsorbable screw for tibial fixation. Mycobacterium fortuitum wasidentified as the pathogen after specific mycobacterial cultures were obtained from operative specimens.M. fortuitum is a known but rare cause of periprosthetic infection. Diagnosis is often delayed as routinemicrobiological cultures do not utilise specific culture requirements for mycobacterial growth. There havebeen several reports in the literature of sterile abscesses associated with bioabsorbable screws. To ourknowledge, this is the first case report of a non-tuberculous mycobacterial infection associated with abioabsorbable implant. This case illustrates that post-operative Mycobacterium infection can occur as acomplication of ACL reconstruction with bioabsorbable screw fixation and should be considered in thedifferential diagnosis of post-operative periprosthetic infection.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

Bioabsorbable interference screws are commonly used forfixation inanterior cruciate ligament (ACL) reconstruction. Reported advantagesover metallic interference screws include lower chance of graftlaceration, lack of image distortion on radiological imaging, anddecreased need for removal in revision surgery [1,2]. Polymersused for bioabsorbable screws include poly-L-lactic acid (PLLA), poly-D,L-lactic acid (PDLLA), polyglycolic acid (PGA), polylactide carbonate,and polylactic acid combined with tricalcium phosphate as in the Bio-INTRAFIX (DePuy Mitek, Raynham, MA) screw used in this case. Therehave been several reports of granulomatous reactions, sterile abscessesand cyst formation associated with bioabsorbable implants [1–5].Routine microbiological cultures in these reports were negative.

We report a case ofMycobacterium fortuitum infection as a cause of anon-healing discharging sinus tract associatedwith a tibial polylactic acidtricalciumphosphate screw and sheath. To our knowledge, this is thefirstcase report of a non-tuberculous mycobacterial infection associated witha bioabsorbable implant. We aim to highlight the similar clinicalpresentation of this case to such sterile reactions recently reported, andemphasise the need to exclude mycobacterial periprosthetic infection.

2. Case report

A15year old girl presentedwith recurrent left knee instabilitywhilstplaying soccer following a non-contact pivoting injury 10months prior.

+61 2 8307 0334.u (S.J. MacDessi).

ll rights reserved.

Clinical and MRI findings showed a displaced bucket handle tear of themedial meniscus, in addition to a full-thickness tear of the ACL. Anarthroscopic four-strand autologous hamstring ACL reconstruction andmedial meniscal repair was performed. The graft was secured with anEndobutton (Smith andNephew, Andover,MA) on the femoral side anda Bio-INTRAFIX sheath and screw on the tibial side.

Post-operative recovery was uneventful at clinic reviews at both 2and 6 weeks. The patient then traveled overseas to Hong Kong for1 month. She was reviewed on her return at post-operative week 11.The tibial wound had developed a 10 mm area of dehiscence distallywith yellow exudate. Oral cephalexin was commenced. There were nosigns of a deep infection with no knee joint effusion and a rangeof motion from 5 to 90°. Routine blood indices including an ESR andC-reactive proteinwerewithinnormal limits. ESRwas14mm/handCRP1.2 mg/L. Exploration in the operating room revealed a superficialwound dehiscence with no apparent communication with the tibialtunnel. The wound was debrided and primarily closed. Cephalexin wascontinuedpost-operatively. Intra-operativemicrobiological samples didnot grow any organisms.

On review 1 week later, a small amount of clear yellow exudatecontinued. Antibiotics were changed to amoxycillin with clavulanicacid. A swab of the exudate again did not reveal any bacterial growth.The patient was reviewed weekly for another month, during whichtime the exudate persisted. Roxithromycin was then added forbroader antimicrobial coverage. A repeat MRI, performed 4 monthsafter the initial surgery, showed diffuse soft tissue swelling and fluidcollection around the tibial screw and tunnel (Figs. 1 and 2). The ACLreconstruction showed normal post-operative high intra-substancesignal and the medial meniscus repair was intact.

Fig. 1. Coronal T-2 weighted magnetic resonance image at 4 months after ACLreconstruction. Moderate fluid is seen surrounding the entire tibial tunnel associatedwith diffuse marrow oedema.

177H.L. Oh et al. / The Knee 17 (2010) 176–178

The patient returned to the operating room and a 5 mm wide sinustract from the tibial wound was found to communicate with the tibialtunnel. The Bio-INTRAFIX screw and sheath were loose and easilyremoved. Clear fluid was noted around the implant. A formal curettageof the tibial tunnel was performed and a V.A.C. dressing was applied tothewound. Oneweek later, thewound basewas cleanwith no evidenceof residual infection and delayed primary closure of the wound wasperformed.

On this occasion, operative samples were tested for mycobacte-rium in addition to routine microbiologic cultures. This yielded a

Fig. 2. Sagittal T-2 weighted magnetic resonance image at 4 months after ACLreconstruction. Extensive soft tissue swelling is seen in the anterior medial aspect aroundthe screw end with fluid after intravenous gadolinium contrast.

heavy growth of M. fortuitum. Based on antibiotic susceptibilities, thepatient was treated with linezolid and moxifloxacin. Linezolid wasceased after 10 weeks and moxifloxacin was continued as mono-therapy for a total of 6 months. The wound successfully healed at7 months after the initial surgery. The patient achieved a full range ofknee movement and stability was maintained with a negativeLachman's test and a negative pivot shift test. She did not have anyremaining knee symptoms and achieved a return to pivoting sports.

3. Discussion

Polyglycolic acid implants causing local inflammatory and osteolyticforeign-body reactions occur in 5–11%of patients, and often present as alate draining sinus secondary to sterile abscess formation [3]. This hasbeen attributed to the rapid degradation of the polyglycolic acid. Localswelling and an inflammatory reaction is thought to be less commonwithpolylactic acid implants as the degradation time is several years [3].Martinek and Friederich reported the first case of local inflammatoryreaction to a bioabsorbable interference screw after ACL reconstructionin 1999. A pretibial cyst continuous with an enlarged tibial tunnel wasfound8monthsafter ACL reconstructionfixedwitha PDLLAscrew in thetibial tunnel [4]. Busfield andAnderson reported in 2007 on two cases ofsterile pretibial abscesses with minimal osteolysis at 3 years and18 months after ACL reconstruction with PLLA interference screws. Inboth these cases, an abscess cavity was found in communication withthe tibial tunnel [5]. Gram stains and cultures were negative. Morerecently, in 2008, Dujardin et al. reported on a patient with a tibial cystand intra-articular granuloma 6 months after ACL reconstruction fixedwith a polylactide carbonate screw [1]. Operative Gram stains andcultures were also negative.

M. fortuitum is an uncommon but known cause of surgical woundand periprosthetic infection [6–9]. It is a rapidly growing non-tuberculous mycobacterium, which exists in soil, domestic watersupplies and has been noted in surgical procedures where irrigationfluids are used. Atypical mycobacterial periprosthetic infection maypresent with fever, pain, swelling, erythema, sinus tract formation, orpurulent drainage [6]. Tissue reaction can be pyogenic or granuloma-tous. Diagnosis is commonly delayed as routinemicrobiological culturesoften do not utilise specific culture requirements for mycobacterialgrowth [6–9].

Atypical mycobacterial infection can be the cause of a dischargingsinus tract associated with a bioabsorbable screw. Diagnosis wasdelayed in our case until mycobacterial cultures were specificallyrequested. We suggest that in all cases of sinuses and abscessesassociated with bioabsorbable implants, specimens should be sent formycobacterial culture in addition to routine microbiological culture.Inflammatory changes with negative routine cultures should only beattributed to reaction to the bioabsorbable screw if atypical mycobac-terial infection is also excluded.

Acknowledgements

The authors would like to acknowledge Dr Michael Houang fromSydney CT & MR for the images provided.

References

[1] Dujardin J, Vandenneucker H, Bellemans J. Tibial cyst and intra-articular granulomaformation after anterior cruciate ligament reconstruction using polylactidecarbonate osteoconductive interference screws. Arthroscopy 2008;24:238–42.

[2] Kwak JH, Sim JA, Kim SH, Lee KC, Lee BK. Delayed intra-articular inflammatoryreaction due to poly-L-lactide bioabsorbable interference screw used in anteriorligament reconstruction. Arthroscopy 2008;24:243–6.

[3] Mosier-LaClair S, PikeH, PomeroyG. Intraosseous bioabsorbable poly-L-lactic acid screwpresenting as a late foreign-body reaction: a case report. FootAnkle Int 2001;22:247–51.

[4] Martinek V, Friederich NF. Tibial and pretibial cyst formation after anterior cruciateligament reconstruction with bioabsorbable interference screw fixation. Arthroscopy1999;15:317–20.

178 H.L. Oh et al. / The Knee 17 (2010) 176–178

[5] Busfield BT, Anderson LJ. Sterile pretibial abscesses after anterior cruciate ligamentreconstruction from bioabsorbable interference screws: a report of 2 cases.Arthroscopy 2007;23:911.e1–4 [Available online at www.arthroscopyjournal.org].

[6] Porat MD, Austin MS. Bilateral knee periprosthetic infection with Mycobacteriumfortuitum. J Arthroplasty 2008;23:787–9.

[7] Herold RC, Lotke PA, MacGregor RR. Prosthetic joint infections secondary to rapidlygrowing Mycobacterium fortuitum. Clin Orthop Relat Res 1987;216:183–6.

[8] Jacoby SM, Sivalingam JJ, Raikin SM. Mycobacterium fortuitum infection followingprimary Achilles tendon debridement with flexor hallucis longus augmentation: acase report. Foot Ankle Int 2008;29:538–41.

[9] Cheung I, Wilson A. Mycobacterium fortuitum infection following total kneearthroplasty: a case report and literature review. Knee 2008;15:61–3.