SURGICAL INFECTIONS Volume 15, Number 5, 2014 ª Mary Ann Liebert, Inc. DOI: 10.1089/sur.2012.133
Case Reports
Successful Two-Stage Revision of Lactobacillus Infection of a Total Knee Arthroplasty and Literature Review Nathan Jacobson, Michael Milshteyn, and Robert Teitge
Abstract
Background: Lactobacillus has been identified as the causative organism in only two total joint arthroplasties: one total hip arthroplasty (THA) and one total knee arthroplasty (TKA). The THA was a litigious case that was treated successfully with a one-stage revision, and the patient with the TKA ultimately required above-knee amputation. We present the first case report of a Lactobacillus infection of a TKA that was treated successfully with a two-stage revision TKA. Methods: Case report and literature review. Case Report: While undergoing a revision left TKA for persistent pain, a 55-year-old Caucasian female was found to have a Lactobacillus infection by intra-operative culture. After an extended course of intravenous (IV) antibiotics, the patient underwent a successful two-stage revision of her prosthesis. Conclusion: Lactobacillus has been appearing more frequently in the literature as a pathogen in patients with compromised immune systems. Orthopedic implant infection with Lactobacillus has only been reported twice, with poorly defined treatment regimens and inconsistent results in both cases. The present report provides orthopedic surgeons and infectious diseases specialists with a treatment algorithm consisting of a two-stage revision of a TKA and a second-line IV antibiotic regimen that may be able to eradicate a Lactobacillus infection of an orthopedic prosthesis with retention of the extremity containing the implant and re-implantation of a functioning prosthesis.
L
actobacillus is a non-spore–forming, microaerophilic, gram-positive rod that ferments glucose to lactic acid and is part of the normal flora of the mucosa of the human oral cavity and gastrointestinal (GI) and genitourinary (GU) tracts [1, 2]. It has recently been cited with increasing frequency in the literature as a pathologic organism, with the majority of cases of Lactobacillus infection occurring in persons with compromised immune systems, including those with diabetes mellitus, malignant disease, and neutropenia, persons undergoing immunosuppressive therapy and broadspectrum antibiotic therapy, as well as patients who have undergone recent operations on the GI tract [2]. Lactobacillus casei and L. rhamnosus have been the most common species of Lactobacillus identified in human oral, GI, and GU infections, but the literature does not routinely describe the identification or differentiation of species of the organism. Lactobacillus is reported almost exclusively as a rare cause of infective endocarditis [3–5] and of infections of vascular grafts and cardiac valve prostheses [1,6]. The literature contains only four case reports of Lactobacillus as the causative organism of
othopedic infections, including two native hip infections [7,8] and two infections of prostheses (one total hip arthroplasty [THA] and one total knee arthroplasty [TKA]) [5,9]. None of the case reports of orthopedic infections with Lactobacillus were found in the orthopedic literature. Transient bacteremia has been described as the source of bacterial seeding of prostheses leading to infection and prosthesis loss [10], and is associated with many diagnostic and therapeutic procedures performed commonly, resulting in late prosthetic joint infection in 0.3%–1.0% of cases [1,11–15]. Bacteremia has been associated with dental procedures [8,14,16], endoscopy [13,15,17], GI surgery [18], GU surgery [19], and head and neck surgery [4]. With the increased utilization of minimally invasive screening procedures such as colonoscopy, cystoscopy, and esophagogastroduodenoscopy, the number of possible episodes of bacteremia than an individual patient may experience is increasing. Infection of a prosthesis generally requires complete removal of the hardware of the prosthetic device and a staged revision. Re-implantation of a prosthesis is done only after
Department of Orthopaedics, Wayne State University School of Medicine, Detroit, Michigan.
635
636
JACOBSON ET AL.
the infection has been cleared with the combination of an antibiotic-eluting spacer and intravenous (IV) antibiotic therapy [10,20,21]. The two Lactobacillus infections reported in the literature consisted of a THA treated with a single-stage revision and of a TKA that ultimately required above-knee amputation [8,9]. In contrast to these two previous case reports, the present case presentation constitutes the first reference in the literature to a TKA infected with Lactobacillus and treated successfully, with retention of the prosthetized limb, as well as the first description of a Lactobacillus-infected total joint arthroplasty treated successfully with a two-stage revision. The patient on whom the following report is based provided informed consent to publication of the findings in her case. No funding or sponsorship was obtained for the preparation of this article. Case Report A 55-year–old female who worked as a business professional had a protracted and complicated surgical history with regard to her left knee, beginning with a rotational injury that occurred during downhill skiing. The patient’s extensive list of surgical interventions on her left knee, described in chronologic order, showing in parentheses the number of year(s) at which each procedure was performed before the index procedure (fourth revision), were: (1) Resection of the medial and lateral meniscus and lateral patellofemoral ligament release ( > 20 yrs, done at another facility); (2) reconstruction of the lateral patellofemoral ligament of the left knee (18 yrs); (3) osteotomy for proximal tibial valgus (8 yrs); (4) left TKA (4 years); and (5) three revisions of the tibial component of her left TKA (3 yrs, 2 yrs, and 1 yr, respectively), all done for persistent pain or malalignment. As part of the routine pre-operative planning for the patient’s fourth revision of her TKA, a urine culture was done 2 wks before her surgery and was positive for Lactobacillus. The patient was then treated with nitrofurantoin for 4 d of a 7-d course. On the fourth day she developed urticaria and the course of antibiotic was stopped. A repeat urine culture was then performed and was negative, and no further treatment
was initiated. The patient had no symptoms of a urinary tract infection (UTI) at this time, and reported that her knee pain, swelling, and erythema were stable, and that she had no fever or chills. At the time of the patient’s fourth revision, a small, grossly purulent fluid collection (approximately 1 mL) was discovered intra-operatively in the joint capsule of her left knee. On the basis of this discovery, the patient underwent resection of her left TKA with placement of a tobramycin antibiotic spacer (containing 1 g tobramycin per 40 mg of cement). A prophylactic IV dose of 1 g vancomycin was infused before the incision was made for resection, and no tourniquet or Foley catheter was used during the procedure. The results of the patient’s laboratory, pathology, and microbiology investigations are summarized in Table 1. Intra-operative tissue biopsy and histologic examination of a frozen section of tissue from the joint capsule and periarticular tissue of her left knee demonstrated dense lymphocyte infiltrates with > 10 polymorphonuclear neutrophils (PMN) per high power field (HPF) in all fields examined. However; all four intraoperative cultures from the patient’s abscess were negative. Her erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) concentration were obtained from blood work that was done preoperatively. The ESR was within normal limits at 19 mm/h (normal, 0–30 mm/h) whereas the CRP was elevated at 58.8 mg/L (normal, < 9 mg/L) according to the standards used in our institution’s laboratory. The patient was then given a regimen of clindamycin 600 mg IV q8h during her hospital stay and for 6 wks postoperatively. Clindamycin was chosen because of the patient’s multiple medication allergies resulting in rash (clarithromycin, lincomycin, moxifloxacin, meperidine, cephalexin, codeine/ pseudoephedrine/guaiafenesin [Robitussin], codeine, morphine, nitrofurantoin), and anaphylaxis with penicillin. The patient returned for removal of the antibiotic spacer in her left knee and revision of her left TKA at 4 mo after placement of the spacer and 3 wks after completing her 6-wk course of clindamycin. She reported that the pain, swelling, and erythema in her knee had stabilized after her antibiotic treatment and denied experiencing any fever or chills. Her pre-operative ESR and CRP values were within normal limits
Table 1. Summary of Laboratory, Pathology, and Microbiology Results Laboratory findingsa
Pathology and microbiologyb
Procedure and timing
ESR 0–30 mm/h
CRP < 9 mg/L
PMN per HPF
Culture
2 Wks preoperatively
-
-
-
First spacer placement (time 0)
19
58.8
> 10
Second spacer placement ( + 4 mos)
12
7
> 10
Revision TKA ( + 7 mos)
15
7.8
Urine: Lactobacillus Tissue and fluidc: Negative >Tissue and fluid: Lactobacillus Tissue and fluid: Negative
< 10 in 9/10 specimens
a C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) were measured as markers of infection at each stage of the patient’s surgical treatment. Intra-operative pathology examination of frozen sections was done with an evaluation of > 20 high powered fields (HPF) and tissue cultures were obtained. b Lactobacillus was identified on two cultures, but specific identification and sensitivities were not done because of laboratory limitations. c Tissue and fluid: joint fluid swabs obtained intra-operatively and tissue biopsy specimens. ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; PMN = polymorphonuclear neutrophils; HPF = high-power field; IV = intravenous; PO = per os; TKA = total knee arthroplasty.
LACTOBACILLUS INFECTION OF A TKA before reimplantation (Table 1). However, an intra-operative pathology examination again demonstrated elevated numbers of PMN ( > 10 per HPF and multiple cultures were positive for Lactobacillus (Table 1). Consequently, the patient’s tobramycin-containing antibiotic spacer was revised and she was begun on a new, extended regimen of vancomycin 1.25 g IV q12 h for 6 wks post-operatively. In an attempt to identify a source of the patient’s Lactobacillus infection other than her previous urinary tract infection (UTI), she underwent an orthopantogram (Panorex), retroperitoneal ultrasound (US) examination, two-dimensional (2D) echocardiography, venous US examination, and 12-lead electrocardiography, all of which were negative or within normal limits. On further inquiry into the patient’s past medical history, it was found that she had been diagnosed with dental caries before her original surgery and had not undergone any treatment for this. As a result, the treatment team involved in her revision TKA recommended that before having a revision TKA she seek intervention for her longstanding dental issues. The patient subsequently underwent multiple dental treatments and interventions. Seven months after the patient’s original surgery and the identification of her infection, she underwent successful reimplantation of a left TKA. Her pre-operative ESR and CRP levels were within normal limits and intra-operative cultures were negative (Table 1). Intra-operative frozen-section examinations for tissue pathology of more than 20 HPF identified one of 10 specimens that had 10 PMN in a single HPF. With multiple negative indicators for infection, and a single positive HPF among more than 200 HPF evaluated, the decision was made to proceed with re-implantation and initiation of an extended course of oral antibiotic therapy with sulfamethoxazole/trimethoprim, 160 mg/800mg qd, for 6 mo. After completion of the course of oral antibiotic therapy the patient’s ESR and CRP remained within normal limits. Findings at a follow-up examination 2 yr after re-implantation of her TKA were benign. The patient’s ESR and CRP remained within normal limits and no reason was found for clinical concern about infection. The patient has shown a significant reduction in pain from its baseline level in activities of daily living and low-impact exercise. At a final follow-up examination the range of motion of her prosthetic was 0–100 degrees. Discussion The following sections summarize the two cases of Lactobacillus infections of prostheses previously documented in the literature [8,9]. Case 1: Total hip arthroplasty Marder [8] presented the case of an elderly male with a history of type 2 diabetes mellitus, bipolar disease controlled with lithium, and a long history of dental caries. The patient had a well-functioning left THA before undergoing a maxillary root canal procedure and extraction of an adjacent first pre-molar root tip. No antibiotic prophylaxis was used before the dental procedure and the patient presented 1wk later with left hip pain, which was identified as coming from an infection and was treated with irrigation and hip debridement. Cultures performed at the time of irrigation and debridement were positive for Lactobacillus. The patient subsequently had three dislocations of his left hip before undergoing a
637 successful single-stage revision of his THA. Following this, he filed an unsuccessful lawsuit against the dentist who had done his root canal surgery for malpractice and improper use of prophylactic antibiotics. No further outcomes, history, or complications were described in the case of this patient. Case 2: Total knee arthroplasty Atwal et al. [9] presented the case of an eight-year–old male with a history of hypertension and peripheral vascular disease. The patient had previously undergone two revision TKAs. One revision was for a presumed infection, but all cultures in this procedure were negative. The second revision was for an implant failure that resulted in significant bone loss requiring allografting and reconstruction at the time of the revision. The patient originally presented in a febrile (39.5 C) state with an inflamed and swollen knee joint. Aspiration culture was negative but intra-operative tissue culture demonstrated Lactobacillus paracasei. There was no report of a suspected origin or possible cause of a transient bacteremia. The patient was treated with IV amoxicillin followed by oral clindamycin, and showed initial improvement, but ultimately relapsed after ‘‘a couple of weeks,’’ at which time an aboveknee amputation was performed. The infecting Lactobacillus in the case of our patient presented here most likely seeded her TKA in an episode of transient bacteremia from either her GU tract or dental carries. She had a history of a positive urine culture for which she appeared to have been adequately treated with an abridged course of antibiotics, as demonstrated by negative results in a repeat urinalysis. Foley catheters add to the risk of a UTI and bacteremia [22,23]; however, no catheter was used during the operative procedure (primary TKA) or during the postoperative hospital stay. A retrospective chart review by Koulouvaris et al., with a case-control analysis of 19,735 patients who underwent total joint arthroplasty, found that 58 of the patients (0.29%) developed surgical site infections (42 deep and 16 joint-space infections), and that seven of these 58 patients had a UTI (three preoperatively and four post-operatively) [24]. Only one of the 58 infections demonstrated the same bacteria in both urinary and wound cultures, indicating a minimal incidence of deep wound or joint infection from a UTI [24]. The second possible origin of the Lactobacillus bacteremia in the case of our patient would have been her long-standing dental carries, a condition that has been documented in the literature [8] as posing a potential risk for transient bacteremia, and thereby for seeding her TKA. This was addressed after revision of the patient’s antibiotic spacer in an attempt to eliminate any ongoing modifiable risk factors. A recent review by Cannon et al. [1] well illustrates the pathogenicity of Lactobacillus. The species L. casei and L. rhamnosus were the most common pathologic isolates in their review, and when all species were considered collectively, Lactobacillus was generally sensitive to erythromycin (94.3%) and clindamycin (90.9%) and resistant to vancomycin (22.5%) [1]. The species L. acidophilus was most sensitive to vancomycin (75%). Consequently, erythromycin would appear to be a potential drug of choice for treating Lactobacillus infections in the subset of patients with anaphylactic allergies to penicillin and possible cross-hypersensitivity to betalactam antibiotics such as carbapenems. However, treatment
638 with erythromycin requires the greatest derangement in patient lifestyle of any antibiotic drug, owing to its requirement for QID dosing. By comparison, treatment with clindamycin requires TID dosing and vancomycin requires BID dosing. However, the antibiotic treatment of our patient varied because of her multiple drug allergies. She was initially treated with clindamycin 600 mg IV q8h to optimizate her antibiotic therapy while allowing her to continue to participate in some of her professional duties on a limited basis. After 6 wks of treatment with this regimen her infection had been controlled, as seen with the improvement in her symptoms and normalization of her ESR and CRP concentration. Nevertheless, her infection had not been cleared, as shown by the intra-operative histologic examination that revealed a count of >10 PMN per HPF. Because of pressure from the patient for fewer occupational restrictions, and the failure of her clindamycin therapy, the decision was made at the time of revision of her spacer to change her antibiotic regimen to vancomycin 1.25 g IV BID. As a result of the decision to utilize an antibiotic that carries a higher risk of Lactobacillus resistance (22.5%) [1], the patient required closer monitoring and agreed to submit to bi-weekly evaluation of her ESR and CRP. Her ESR and CRP values demonstrated an appropriate increase in response to the surgical interventions undertaken in the face of her condition, and then became normal, where they remained until the final revision of her TKA. After the patient’s infection had cleared and she had undergone reimplantation of her TKA, lifelong suppressive therapy for her infection was considered with a regimen of sulfamethoxazole/trimethoprim 160 mg/800 mg QD. However at her 1-yr follow-up visit after reimplantation of her prosthesis, the patient informed the treatment team that she had stopped her regimen of sulfamethoxazole/trimethoprim after 6 mo without physician consultation. Although the patient had failed to continue her antibiotic therapy, the findings on her examination were benign, and her ESR and CRP concentration were within normal limits. because of this, the treatment team decided that no additional treatment was necessary. The work of Cannon et al. and Salminen et al. suggest that when Lactobacillus is identified by culture it should not be treated as a contaminant, but should be considered seriously as a pathogen in selecting an antibiotic treatment regimen, especially when identified in multiple cultures, as in the case of our patient described here. Salminen et al. reviewed retrospectively 89 cases of Lactobacillus bacteremia, in which mortality was 26% at 1 mo and 48% at 1 yr [25]. In multivariable analysis, severe underlying diseases were significant predictors of mortality, whereas the associated mortality with antimicrobial treatment was lower [25]. There was a significant association of polymicrobial infection with mortality, with 43.5% of patients with polymicrobial infections dying as opposed to 23.0% of those without [1]. Lactobacillus is a potential cause of atypical infection of total joint arthroplasties through seeding from transient bacteremia associated with dental caries and UTI. If Lactobacillus is identified in an infected total joint it should not be considered a contaminant, and appropriate antibiotic treatment should be administered. The sensitivity of Lactobacillus to erythromycin (94.3%) is higher than that to other antibiotics, and erythromycin should therefore be con-
JACOBSON ET AL. sidered a first-line treatment for Lactobacillus infection, followed by clindamycin (90.9%). If vancomycin must be used because of a patient’s allergies or for significant social reasons and because it allows a more favorable dosing schedule, it should be given with close monitoring and frequent assay of the patient’s ESR and CRP concentration. The case of our patient described here is the first reported case of Lactobacillus infection of a TKA successfully treated with limb retention, as well as being the first case of a Lactobacillusinfected total joint arthroplasty successfully treated with a two-stage revision. Author Disclosure Statement No competing financial interests exist. References 1. Cannon JP, Lee TA, Bolanos JT, Danzinger LH. Pathogenic relevance of Lactobacillus: A retrospective review of over 200 cases. Eur J Clin Microbio Infect Dis 2005;24:31–40 2. Antony SJ. Lactobacillemia: An emerging cause of infection in both the immunocompromised and the immunocompetent host. J Natl Med Assoc 2000; 92:83–86. 3. Salvana EM, Frank M. Lactobacillus endocarditis: Case report and review of cases reported since 1992. J Infect 2006;53: e5–e10. 4. Yagi S, Akaike M, Fujimura M, et al. Infective endocarditis caused by Lactobacillus. Intern Med 2008;47:1113–1116. 5. Presterl E, Kneifel W, Mayer HK, et al. Endocarditis by Lactobacillus rhamnosus due to yogurt ingestion? Scand J Infect Dis 2001;33:710–714. 6. Schoon Y, Schuurman B, Buiting AG, Kranendonk SE, Graafsma SJ. Aortic graft infection by Lactobacillus casei: A case report. Neth J Med 1998;52:71–74. 7. Chanet V, Brazille P, Honore S, et al. Lactobacillus septic arthritis. South Med J. 2007;100:531–532. 8. Marder MZ. Dental treatment and orthopedic implants—A malpractice case. Gen Dent. 2005;53:125–129. 9. Atwal N, George A, Squires B, Marsh CH. Lactobacillus as a rare cause of an infected total knee replacement: A case report. J Med Case Rep 2009;33:7441. 10. McDonald DJ, Fitzgerald RH Jr, Ilstrup DM. Two-stage reconstruction of total hip arthroplasty because of infection. J Bone Joint Surg Am 1989;71:828–834. 11. Ahlberg A, Carlsson AS, Lindberg L. Hematogenous infection in total joint replacement. Clin. Orthop 1978;137:69 –75. 12. Jacobson JJ, Millard HD, Plezia, R, Blankenship JR. Dental treatment and late prosthetic joint infections. Oral Surg Oral Med Oral Pathol 1986, 61:413–417. 13. Vanderhooft JE, Robinson RP. Late infection of a bipolar prosthesis following endoscopy. A case report. J Bone Joint Surg Am 1994;76:744–746. 14. Jacobson JJ, Matthews LS. Bacteria isolated from late prosthetic joint infections: dental treatment and chemoprophylaxis. Oral Surg Oral Med Oral Pathol 1987;63:122–126. 15. Weiler PJ. Late infection of a bipolar prosthesis following endoscopy. A case report J Bone Joint Surg Am 1995;77: 1129–1130. 16. Rajasuo A, Nyfors S, Kanervo A, et al. Bacteremia after plate removal and tooth extraction. Int J Oral Maxillofac Surg 2004;33:356–360. 17. Avlami A, Kordossis T, Vrizidis N, Sipsas NV. Lactobacillus rhamnosus endocarditis complicating colonoscopy. J Infect 2001;42:283–285.
LACTOBACILLUS INFECTION OF A TKA 18. Bayer AS, Chow AW, Betts D, Guze LB. Lactobacillemia— Report of nine cases. Am J Med 1978;64:808–813. 19. Goldberg VM, Henderson BT. The Freeman–Swanson 1CLH total knee arthroplasty. Complications and problems. J Bone Joint Surg Am 1980; 62:1338–1344. 20. Tsukayama DT, Goldberg VM, Kyle R. Diagnosis and management of infection after total knee arthroplasty. J Bone Joint Surg Am 2003;85:S75–S80. 21. Fitzgerald RH. Infected total hip arthroplasty: Diagnosis and treatment. J Am Acad Orthop Surg 1995;3:249– 2620. 22. Stamm WE. Catheter-associated urinary tract infections: epidemiology, pathogenesis, and prevention. Am J Med 1991; 91:65S–71S. 23. Stephan F, Sax H, Wachsmuth M, Hoffmeyer P, Clergue F, Pittet D. Reduction of urinary tract infection and antibiotic use after surgery: a controlled, prospective, before-after intervention study. Clin Infect Dis 2006;42:1544–1551.
639 24. Koulouvaris P, Sculco P, Finerty E, Sculco T, Sharrock NE. Relationship between perioperative urinary tract infection and deep infection after joint arthroplasty. Clin Orthop Relat Res 2009;467:1859–1867. 25. Salminen MK, Rautelin H, Tynkkynen S, et al. Lactobacillus bacteremia, clinical significance and patient outcome, with special focus on probiotic L. rhamonosus GG. Clin Infect Dis 2004;38:62–69.
Address correspondence to: Dr. Nathan Jacobson Department of Orthopaedics Wayne State University School of Medicine Detroit, MI 48202 E-mail: [email protected]
Case Reports
Successful Two-Stage Revision of Lactobacillus Infection of a Total Knee Arthroplasty and Literature Review Nathan Jacobson, Michael Milshteyn, and Robert Teitge
Abstract
Background: Lactobacillus has been identified as the causative organism in only two total joint arthroplasties: one total hip arthroplasty (THA) and one total knee arthroplasty (TKA). The THA was a litigious case that was treated successfully with a one-stage revision, and the patient with the TKA ultimately required above-knee amputation. We present the first case report of a Lactobacillus infection of a TKA that was treated successfully with a two-stage revision TKA. Methods: Case report and literature review. Case Report: While undergoing a revision left TKA for persistent pain, a 55-year-old Caucasian female was found to have a Lactobacillus infection by intra-operative culture. After an extended course of intravenous (IV) antibiotics, the patient underwent a successful two-stage revision of her prosthesis. Conclusion: Lactobacillus has been appearing more frequently in the literature as a pathogen in patients with compromised immune systems. Orthopedic implant infection with Lactobacillus has only been reported twice, with poorly defined treatment regimens and inconsistent results in both cases. The present report provides orthopedic surgeons and infectious diseases specialists with a treatment algorithm consisting of a two-stage revision of a TKA and a second-line IV antibiotic regimen that may be able to eradicate a Lactobacillus infection of an orthopedic prosthesis with retention of the extremity containing the implant and re-implantation of a functioning prosthesis.
L
actobacillus is a non-spore–forming, microaerophilic, gram-positive rod that ferments glucose to lactic acid and is part of the normal flora of the mucosa of the human oral cavity and gastrointestinal (GI) and genitourinary (GU) tracts [1, 2]. It has recently been cited with increasing frequency in the literature as a pathologic organism, with the majority of cases of Lactobacillus infection occurring in persons with compromised immune systems, including those with diabetes mellitus, malignant disease, and neutropenia, persons undergoing immunosuppressive therapy and broadspectrum antibiotic therapy, as well as patients who have undergone recent operations on the GI tract [2]. Lactobacillus casei and L. rhamnosus have been the most common species of Lactobacillus identified in human oral, GI, and GU infections, but the literature does not routinely describe the identification or differentiation of species of the organism. Lactobacillus is reported almost exclusively as a rare cause of infective endocarditis [3–5] and of infections of vascular grafts and cardiac valve prostheses [1,6]. The literature contains only four case reports of Lactobacillus as the causative organism of
othopedic infections, including two native hip infections [7,8] and two infections of prostheses (one total hip arthroplasty [THA] and one total knee arthroplasty [TKA]) [5,9]. None of the case reports of orthopedic infections with Lactobacillus were found in the orthopedic literature. Transient bacteremia has been described as the source of bacterial seeding of prostheses leading to infection and prosthesis loss [10], and is associated with many diagnostic and therapeutic procedures performed commonly, resulting in late prosthetic joint infection in 0.3%–1.0% of cases [1,11–15]. Bacteremia has been associated with dental procedures [8,14,16], endoscopy [13,15,17], GI surgery [18], GU surgery [19], and head and neck surgery [4]. With the increased utilization of minimally invasive screening procedures such as colonoscopy, cystoscopy, and esophagogastroduodenoscopy, the number of possible episodes of bacteremia than an individual patient may experience is increasing. Infection of a prosthesis generally requires complete removal of the hardware of the prosthetic device and a staged revision. Re-implantation of a prosthesis is done only after
Department of Orthopaedics, Wayne State University School of Medicine, Detroit, Michigan.
635
636
JACOBSON ET AL.
the infection has been cleared with the combination of an antibiotic-eluting spacer and intravenous (IV) antibiotic therapy [10,20,21]. The two Lactobacillus infections reported in the literature consisted of a THA treated with a single-stage revision and of a TKA that ultimately required above-knee amputation [8,9]. In contrast to these two previous case reports, the present case presentation constitutes the first reference in the literature to a TKA infected with Lactobacillus and treated successfully, with retention of the prosthetized limb, as well as the first description of a Lactobacillus-infected total joint arthroplasty treated successfully with a two-stage revision. The patient on whom the following report is based provided informed consent to publication of the findings in her case. No funding or sponsorship was obtained for the preparation of this article. Case Report A 55-year–old female who worked as a business professional had a protracted and complicated surgical history with regard to her left knee, beginning with a rotational injury that occurred during downhill skiing. The patient’s extensive list of surgical interventions on her left knee, described in chronologic order, showing in parentheses the number of year(s) at which each procedure was performed before the index procedure (fourth revision), were: (1) Resection of the medial and lateral meniscus and lateral patellofemoral ligament release ( > 20 yrs, done at another facility); (2) reconstruction of the lateral patellofemoral ligament of the left knee (18 yrs); (3) osteotomy for proximal tibial valgus (8 yrs); (4) left TKA (4 years); and (5) three revisions of the tibial component of her left TKA (3 yrs, 2 yrs, and 1 yr, respectively), all done for persistent pain or malalignment. As part of the routine pre-operative planning for the patient’s fourth revision of her TKA, a urine culture was done 2 wks before her surgery and was positive for Lactobacillus. The patient was then treated with nitrofurantoin for 4 d of a 7-d course. On the fourth day she developed urticaria and the course of antibiotic was stopped. A repeat urine culture was then performed and was negative, and no further treatment
was initiated. The patient had no symptoms of a urinary tract infection (UTI) at this time, and reported that her knee pain, swelling, and erythema were stable, and that she had no fever or chills. At the time of the patient’s fourth revision, a small, grossly purulent fluid collection (approximately 1 mL) was discovered intra-operatively in the joint capsule of her left knee. On the basis of this discovery, the patient underwent resection of her left TKA with placement of a tobramycin antibiotic spacer (containing 1 g tobramycin per 40 mg of cement). A prophylactic IV dose of 1 g vancomycin was infused before the incision was made for resection, and no tourniquet or Foley catheter was used during the procedure. The results of the patient’s laboratory, pathology, and microbiology investigations are summarized in Table 1. Intra-operative tissue biopsy and histologic examination of a frozen section of tissue from the joint capsule and periarticular tissue of her left knee demonstrated dense lymphocyte infiltrates with > 10 polymorphonuclear neutrophils (PMN) per high power field (HPF) in all fields examined. However; all four intraoperative cultures from the patient’s abscess were negative. Her erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) concentration were obtained from blood work that was done preoperatively. The ESR was within normal limits at 19 mm/h (normal, 0–30 mm/h) whereas the CRP was elevated at 58.8 mg/L (normal, < 9 mg/L) according to the standards used in our institution’s laboratory. The patient was then given a regimen of clindamycin 600 mg IV q8h during her hospital stay and for 6 wks postoperatively. Clindamycin was chosen because of the patient’s multiple medication allergies resulting in rash (clarithromycin, lincomycin, moxifloxacin, meperidine, cephalexin, codeine/ pseudoephedrine/guaiafenesin [Robitussin], codeine, morphine, nitrofurantoin), and anaphylaxis with penicillin. The patient returned for removal of the antibiotic spacer in her left knee and revision of her left TKA at 4 mo after placement of the spacer and 3 wks after completing her 6-wk course of clindamycin. She reported that the pain, swelling, and erythema in her knee had stabilized after her antibiotic treatment and denied experiencing any fever or chills. Her pre-operative ESR and CRP values were within normal limits
Table 1. Summary of Laboratory, Pathology, and Microbiology Results Laboratory findingsa
Pathology and microbiologyb
Procedure and timing
ESR 0–30 mm/h
CRP < 9 mg/L
PMN per HPF
Culture
2 Wks preoperatively
-
-
-
First spacer placement (time 0)
19
58.8
> 10
Second spacer placement ( + 4 mos)
12
7
> 10
Revision TKA ( + 7 mos)
15
7.8
Urine: Lactobacillus Tissue and fluidc: Negative >Tissue and fluid: Lactobacillus Tissue and fluid: Negative
< 10 in 9/10 specimens
a C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) were measured as markers of infection at each stage of the patient’s surgical treatment. Intra-operative pathology examination of frozen sections was done with an evaluation of > 20 high powered fields (HPF) and tissue cultures were obtained. b Lactobacillus was identified on two cultures, but specific identification and sensitivities were not done because of laboratory limitations. c Tissue and fluid: joint fluid swabs obtained intra-operatively and tissue biopsy specimens. ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; PMN = polymorphonuclear neutrophils; HPF = high-power field; IV = intravenous; PO = per os; TKA = total knee arthroplasty.
LACTOBACILLUS INFECTION OF A TKA before reimplantation (Table 1). However, an intra-operative pathology examination again demonstrated elevated numbers of PMN ( > 10 per HPF and multiple cultures were positive for Lactobacillus (Table 1). Consequently, the patient’s tobramycin-containing antibiotic spacer was revised and she was begun on a new, extended regimen of vancomycin 1.25 g IV q12 h for 6 wks post-operatively. In an attempt to identify a source of the patient’s Lactobacillus infection other than her previous urinary tract infection (UTI), she underwent an orthopantogram (Panorex), retroperitoneal ultrasound (US) examination, two-dimensional (2D) echocardiography, venous US examination, and 12-lead electrocardiography, all of which were negative or within normal limits. On further inquiry into the patient’s past medical history, it was found that she had been diagnosed with dental caries before her original surgery and had not undergone any treatment for this. As a result, the treatment team involved in her revision TKA recommended that before having a revision TKA she seek intervention for her longstanding dental issues. The patient subsequently underwent multiple dental treatments and interventions. Seven months after the patient’s original surgery and the identification of her infection, she underwent successful reimplantation of a left TKA. Her pre-operative ESR and CRP levels were within normal limits and intra-operative cultures were negative (Table 1). Intra-operative frozen-section examinations for tissue pathology of more than 20 HPF identified one of 10 specimens that had 10 PMN in a single HPF. With multiple negative indicators for infection, and a single positive HPF among more than 200 HPF evaluated, the decision was made to proceed with re-implantation and initiation of an extended course of oral antibiotic therapy with sulfamethoxazole/trimethoprim, 160 mg/800mg qd, for 6 mo. After completion of the course of oral antibiotic therapy the patient’s ESR and CRP remained within normal limits. Findings at a follow-up examination 2 yr after re-implantation of her TKA were benign. The patient’s ESR and CRP remained within normal limits and no reason was found for clinical concern about infection. The patient has shown a significant reduction in pain from its baseline level in activities of daily living and low-impact exercise. At a final follow-up examination the range of motion of her prosthetic was 0–100 degrees. Discussion The following sections summarize the two cases of Lactobacillus infections of prostheses previously documented in the literature [8,9]. Case 1: Total hip arthroplasty Marder [8] presented the case of an elderly male with a history of type 2 diabetes mellitus, bipolar disease controlled with lithium, and a long history of dental caries. The patient had a well-functioning left THA before undergoing a maxillary root canal procedure and extraction of an adjacent first pre-molar root tip. No antibiotic prophylaxis was used before the dental procedure and the patient presented 1wk later with left hip pain, which was identified as coming from an infection and was treated with irrigation and hip debridement. Cultures performed at the time of irrigation and debridement were positive for Lactobacillus. The patient subsequently had three dislocations of his left hip before undergoing a
637 successful single-stage revision of his THA. Following this, he filed an unsuccessful lawsuit against the dentist who had done his root canal surgery for malpractice and improper use of prophylactic antibiotics. No further outcomes, history, or complications were described in the case of this patient. Case 2: Total knee arthroplasty Atwal et al. [9] presented the case of an eight-year–old male with a history of hypertension and peripheral vascular disease. The patient had previously undergone two revision TKAs. One revision was for a presumed infection, but all cultures in this procedure were negative. The second revision was for an implant failure that resulted in significant bone loss requiring allografting and reconstruction at the time of the revision. The patient originally presented in a febrile (39.5 C) state with an inflamed and swollen knee joint. Aspiration culture was negative but intra-operative tissue culture demonstrated Lactobacillus paracasei. There was no report of a suspected origin or possible cause of a transient bacteremia. The patient was treated with IV amoxicillin followed by oral clindamycin, and showed initial improvement, but ultimately relapsed after ‘‘a couple of weeks,’’ at which time an aboveknee amputation was performed. The infecting Lactobacillus in the case of our patient presented here most likely seeded her TKA in an episode of transient bacteremia from either her GU tract or dental carries. She had a history of a positive urine culture for which she appeared to have been adequately treated with an abridged course of antibiotics, as demonstrated by negative results in a repeat urinalysis. Foley catheters add to the risk of a UTI and bacteremia [22,23]; however, no catheter was used during the operative procedure (primary TKA) or during the postoperative hospital stay. A retrospective chart review by Koulouvaris et al., with a case-control analysis of 19,735 patients who underwent total joint arthroplasty, found that 58 of the patients (0.29%) developed surgical site infections (42 deep and 16 joint-space infections), and that seven of these 58 patients had a UTI (three preoperatively and four post-operatively) [24]. Only one of the 58 infections demonstrated the same bacteria in both urinary and wound cultures, indicating a minimal incidence of deep wound or joint infection from a UTI [24]. The second possible origin of the Lactobacillus bacteremia in the case of our patient would have been her long-standing dental carries, a condition that has been documented in the literature [8] as posing a potential risk for transient bacteremia, and thereby for seeding her TKA. This was addressed after revision of the patient’s antibiotic spacer in an attempt to eliminate any ongoing modifiable risk factors. A recent review by Cannon et al. [1] well illustrates the pathogenicity of Lactobacillus. The species L. casei and L. rhamnosus were the most common pathologic isolates in their review, and when all species were considered collectively, Lactobacillus was generally sensitive to erythromycin (94.3%) and clindamycin (90.9%) and resistant to vancomycin (22.5%) [1]. The species L. acidophilus was most sensitive to vancomycin (75%). Consequently, erythromycin would appear to be a potential drug of choice for treating Lactobacillus infections in the subset of patients with anaphylactic allergies to penicillin and possible cross-hypersensitivity to betalactam antibiotics such as carbapenems. However, treatment
638 with erythromycin requires the greatest derangement in patient lifestyle of any antibiotic drug, owing to its requirement for QID dosing. By comparison, treatment with clindamycin requires TID dosing and vancomycin requires BID dosing. However, the antibiotic treatment of our patient varied because of her multiple drug allergies. She was initially treated with clindamycin 600 mg IV q8h to optimizate her antibiotic therapy while allowing her to continue to participate in some of her professional duties on a limited basis. After 6 wks of treatment with this regimen her infection had been controlled, as seen with the improvement in her symptoms and normalization of her ESR and CRP concentration. Nevertheless, her infection had not been cleared, as shown by the intra-operative histologic examination that revealed a count of >10 PMN per HPF. Because of pressure from the patient for fewer occupational restrictions, and the failure of her clindamycin therapy, the decision was made at the time of revision of her spacer to change her antibiotic regimen to vancomycin 1.25 g IV BID. As a result of the decision to utilize an antibiotic that carries a higher risk of Lactobacillus resistance (22.5%) [1], the patient required closer monitoring and agreed to submit to bi-weekly evaluation of her ESR and CRP. Her ESR and CRP values demonstrated an appropriate increase in response to the surgical interventions undertaken in the face of her condition, and then became normal, where they remained until the final revision of her TKA. After the patient’s infection had cleared and she had undergone reimplantation of her TKA, lifelong suppressive therapy for her infection was considered with a regimen of sulfamethoxazole/trimethoprim 160 mg/800 mg QD. However at her 1-yr follow-up visit after reimplantation of her prosthesis, the patient informed the treatment team that she had stopped her regimen of sulfamethoxazole/trimethoprim after 6 mo without physician consultation. Although the patient had failed to continue her antibiotic therapy, the findings on her examination were benign, and her ESR and CRP concentration were within normal limits. because of this, the treatment team decided that no additional treatment was necessary. The work of Cannon et al. and Salminen et al. suggest that when Lactobacillus is identified by culture it should not be treated as a contaminant, but should be considered seriously as a pathogen in selecting an antibiotic treatment regimen, especially when identified in multiple cultures, as in the case of our patient described here. Salminen et al. reviewed retrospectively 89 cases of Lactobacillus bacteremia, in which mortality was 26% at 1 mo and 48% at 1 yr [25]. In multivariable analysis, severe underlying diseases were significant predictors of mortality, whereas the associated mortality with antimicrobial treatment was lower [25]. There was a significant association of polymicrobial infection with mortality, with 43.5% of patients with polymicrobial infections dying as opposed to 23.0% of those without [1]. Lactobacillus is a potential cause of atypical infection of total joint arthroplasties through seeding from transient bacteremia associated with dental caries and UTI. If Lactobacillus is identified in an infected total joint it should not be considered a contaminant, and appropriate antibiotic treatment should be administered. The sensitivity of Lactobacillus to erythromycin (94.3%) is higher than that to other antibiotics, and erythromycin should therefore be con-
JACOBSON ET AL. sidered a first-line treatment for Lactobacillus infection, followed by clindamycin (90.9%). If vancomycin must be used because of a patient’s allergies or for significant social reasons and because it allows a more favorable dosing schedule, it should be given with close monitoring and frequent assay of the patient’s ESR and CRP concentration. The case of our patient described here is the first reported case of Lactobacillus infection of a TKA successfully treated with limb retention, as well as being the first case of a Lactobacillusinfected total joint arthroplasty successfully treated with a two-stage revision. Author Disclosure Statement No competing financial interests exist. References 1. Cannon JP, Lee TA, Bolanos JT, Danzinger LH. Pathogenic relevance of Lactobacillus: A retrospective review of over 200 cases. Eur J Clin Microbio Infect Dis 2005;24:31–40 2. Antony SJ. Lactobacillemia: An emerging cause of infection in both the immunocompromised and the immunocompetent host. J Natl Med Assoc 2000; 92:83–86. 3. Salvana EM, Frank M. Lactobacillus endocarditis: Case report and review of cases reported since 1992. J Infect 2006;53: e5–e10. 4. Yagi S, Akaike M, Fujimura M, et al. Infective endocarditis caused by Lactobacillus. Intern Med 2008;47:1113–1116. 5. Presterl E, Kneifel W, Mayer HK, et al. Endocarditis by Lactobacillus rhamnosus due to yogurt ingestion? Scand J Infect Dis 2001;33:710–714. 6. Schoon Y, Schuurman B, Buiting AG, Kranendonk SE, Graafsma SJ. Aortic graft infection by Lactobacillus casei: A case report. Neth J Med 1998;52:71–74. 7. Chanet V, Brazille P, Honore S, et al. Lactobacillus septic arthritis. South Med J. 2007;100:531–532. 8. Marder MZ. Dental treatment and orthopedic implants—A malpractice case. Gen Dent. 2005;53:125–129. 9. Atwal N, George A, Squires B, Marsh CH. Lactobacillus as a rare cause of an infected total knee replacement: A case report. J Med Case Rep 2009;33:7441. 10. McDonald DJ, Fitzgerald RH Jr, Ilstrup DM. Two-stage reconstruction of total hip arthroplasty because of infection. J Bone Joint Surg Am 1989;71:828–834. 11. Ahlberg A, Carlsson AS, Lindberg L. Hematogenous infection in total joint replacement. Clin. Orthop 1978;137:69 –75. 12. Jacobson JJ, Millard HD, Plezia, R, Blankenship JR. Dental treatment and late prosthetic joint infections. Oral Surg Oral Med Oral Pathol 1986, 61:413–417. 13. Vanderhooft JE, Robinson RP. Late infection of a bipolar prosthesis following endoscopy. A case report. J Bone Joint Surg Am 1994;76:744–746. 14. Jacobson JJ, Matthews LS. Bacteria isolated from late prosthetic joint infections: dental treatment and chemoprophylaxis. Oral Surg Oral Med Oral Pathol 1987;63:122–126. 15. Weiler PJ. Late infection of a bipolar prosthesis following endoscopy. A case report J Bone Joint Surg Am 1995;77: 1129–1130. 16. Rajasuo A, Nyfors S, Kanervo A, et al. Bacteremia after plate removal and tooth extraction. Int J Oral Maxillofac Surg 2004;33:356–360. 17. Avlami A, Kordossis T, Vrizidis N, Sipsas NV. Lactobacillus rhamnosus endocarditis complicating colonoscopy. J Infect 2001;42:283–285.
LACTOBACILLUS INFECTION OF A TKA 18. Bayer AS, Chow AW, Betts D, Guze LB. Lactobacillemia— Report of nine cases. Am J Med 1978;64:808–813. 19. Goldberg VM, Henderson BT. The Freeman–Swanson 1CLH total knee arthroplasty. Complications and problems. J Bone Joint Surg Am 1980; 62:1338–1344. 20. Tsukayama DT, Goldberg VM, Kyle R. Diagnosis and management of infection after total knee arthroplasty. J Bone Joint Surg Am 2003;85:S75–S80. 21. Fitzgerald RH. Infected total hip arthroplasty: Diagnosis and treatment. J Am Acad Orthop Surg 1995;3:249– 2620. 22. Stamm WE. Catheter-associated urinary tract infections: epidemiology, pathogenesis, and prevention. Am J Med 1991; 91:65S–71S. 23. Stephan F, Sax H, Wachsmuth M, Hoffmeyer P, Clergue F, Pittet D. Reduction of urinary tract infection and antibiotic use after surgery: a controlled, prospective, before-after intervention study. Clin Infect Dis 2006;42:1544–1551.
639 24. Koulouvaris P, Sculco P, Finerty E, Sculco T, Sharrock NE. Relationship between perioperative urinary tract infection and deep infection after joint arthroplasty. Clin Orthop Relat Res 2009;467:1859–1867. 25. Salminen MK, Rautelin H, Tynkkynen S, et al. Lactobacillus bacteremia, clinical significance and patient outcome, with special focus on probiotic L. rhamonosus GG. Clin Infect Dis 2004;38:62–69.
Address correspondence to: Dr. Nathan Jacobson Department of Orthopaedics Wayne State University School of Medicine Detroit, MI 48202 E-mail: [email protected]
