Case Report: Prosthetic knee joint infections as a complication of lumbar osteomyelitis-discitis

Sagar Patel BS, George El-Bahri DO, Bernard D Fishalow MD FAAOS, Patricia Baumann DO,  FAOAO
Bay Pines VA Healthcare system
Bay Pines, FL

Prosthetic joint infection (PJI) is one of the most common and devastating complications of total knee arthroplasty. It is characterized as early, delayed, or late onset. Common treatment options for PJI include debridement, antibiotics, irrigation, retention (DAIR), one-stage revision, two-stage revision. There is a 19-43% recurrence rate based on treatment option one.

Cases of concurrent discitis-osteomyelitis and PJI have not been well documented. We present two cases of osteomyelitis-discitis complicated by PJI. The first involves a 55-year-old male with a history of IV drug abuse (IVDA) and hepatitis C presenting with late onset bilateral PJI and lumbar osteomyelitis-discitis caused by methicillin sensitive Staphylococcus aureus (MSSA). This patient was treated with 18 weeks of anti-staphylococcal antibiotic therapy as well as a two-stage revision of the right knee. The left knee remains to be reimplanted. The second is a 65-year old male with a history of diabetes mellitus, hypertension, chronic kidney disease, and multiple discharges against medical advice who presented with lumbar osteomyelitis-discitis complicated by methicillin resistant Staphylococcus aureus (MRSA) PJI. This patient’s definitive treatment was six weeks of IV daptomycin and remains to be reimplanted.

Currently, arthroplasty is utilized for knee, hip, ankle, shoulder, and wrist joint replacements of which total knee replacements are the most common, with over 600,000 cases performed yearly in the U.S.2. The most common indication for total knee arthroplasty (TKA) is osteoarthritis that can no longer be controlled by medical management and results in inability to perform activities of daily living. The most common complications of TKA are prosthetic joint infection (PJI), deep venous thrombosis, and early prosthesis failure with PJI being the greatest cause of failure after TKA.

Prosthetic joint infection is characterized as either early onset (<3 months postoperative), delayed onset (3-12 months postoperative) or late onset (>12 months postoperatively).  Early onset is usually the result of intraoperative seeding and is usually caused by virulent organisms such as S. aureus, gram negative bacilli, anaerobes, or mixed infections. Delayed onset is also usually due to intraoperative seeding but is caused by less virulent bacteria, such as coagulase-negative staphylococci or enterococci. Late onset is usually caused by hematogenous spread of S.aureus, beta hemolytic streptococci, or enterobacter3.

The Kaplan-Meier risk of PJI is 0.77% at 1 year and 1.58% at six years. Significant risk factors for PJI include age, Charlson comorbidity score, race, and gender1.  It is recommended that ESR and CRP be tested in patients being assessed for possible PJI. Additionally, joint aspiration for culture (aerobic,  anaerobic, fungal, AFB), WBC count and differential should be performed for periprosthetic knee infections with abnormal ESR or CRP4. Frozen sections of peri-implant tissue should be taken in patients undergoing reoperation in which PJI has not been established or excluded as well as multiple cultures before initiating antibiotic treatment4.

Management options of PJI include incision and drainage (I&D), I&D and liner exchange, one-stage revision, two-stage revision, and amputation. Rate of recurrence of PJI is 26% after first-line treatment1. I&D as a first-line treatment had the highest risk of PJI recurrence at 28% at one year and 43% at six years1. Two-stage reimplantation had the highest success rate with only 19% recurrence1.

Most cases of recurrent PJI are not caused by the same organism with approximately 31% occurring with identical organisms. Positive cultures at reimplantation and poor health status are associated with higher rates of recurrence with new organisms5. Strategies to prevent PJI include incorporation of antibiotic loaded bone cement, commonly with vancomycin. PJI is the most common indication for revision TKA and the third most common reason for revision total hip arthroplasty6. Patients receiving treatment for PJI have a two-fold increase in inpatient mortality for each surgical admission compared to aseptic revisions7 (0.77% vs 0.38%).

Our aim is to report two cases of osteomyelitis-discitis complicated by subsequent PJI. The first case describes a 55-year-old man with possible endocarditis and ensuing discitis-osteomyelitis leading to bilateral prosthetic joint infection. The second case is a 65-year-old male with lumbar osteomyelitis-discitis resulting in unilateral prosthetic knee infection. These cases call to attention uncommon complications of bacteremia and osteomyelitis-discitis in patients with prosthetic joints.

Case Report
This is a 55-year-old male who had right total knee arthroplasty in October 2006 followed by a left total knee arthroplasty in March 2010, both performed at our VA facility.  He had a history of multiple knee surgeries dating back to the 1980’s due to traumatic injuries sustained during military service for which records are unavailable.

The patient has a history of intravenous drug abuse (IVDA), alcohol abuse with chronic pancreatitis, multiple psychiatric conditions and Hepatitis C, with complications including left dorsal hand cellulitis treated with incision and drainage (I&D) in December 2011 and a right upper extremity abscess in April 2014. In January 2015, he was admitted to an outside medical facility for back pain and MSSA bacteremia with possible endocarditis. On March 18, 2015, he was seen in the emergency department for low back pain and abdominal pain and was admitted with chronic pancreatitis.

Upon initial presentation, the patient was started on vancomycin and zosyn empirically. During his hospital stay, orthopedics was consulted for bilateral knee pain and swelling and back pain. On March 23, 2015 MRI of the lumbar spine showed osteomyelitis/discitis at L2-L3 and L3-L4. The knee joint was aspirated on March 24 2015 and was positive for MSSA. The right knee implant was removed, with I&D and insertion of an antibiotic spacer (Cobalt G-Hv™ bone cement loaded with two grams vancomycin per bag of cement) on March 31, 2015. Antibiotics were changed to nafcillin 2gm q4hr on March 24, 2015 after susceptibilities were obtained.

The patient additionally had osteomyelitis of the scaphoid. During his hospital stay the patient also complained of left knee pain. On April 10, 2015 the left knee implant was also removed with insertion of an antibiotic spacer. His regimen of nafcillin was continued for 14 weeks. Response was tracked via ESR, CRP and WBC count (shown below). On June 7, 2015 he was readmitted for possible sepsis and initiated on vancomycin and ceftriaxone, which was subsequently changed to vancomycin and cefepime with blood cultures showing bacillus.

On June 30, 2015 nafcillin IV was changed to dicloxacillin 500mg q6hr PO, which was continued for two weeks. On July 14, 2015 cefipime and vancomycin were discontinued.  The patient underwent a total of 18 weeks of anti-staphylococcal antibiotic therapy including 14 weeks of nafcillin. The patient was seen in the office continuously and the left knee was reaspirated on August 17, 2016 and the right knee was aspirated November 29, 2016 with negative culture. On February 15, 2017 pathology specimens from the right anterior femur, femoral canal, and tibial canal showed significant neutrophils. Neutrophils were not seen in the right posterior knee. The patient underwent reimplantation of right total knee arthroplasty at that time.

Patel graphs

Case 2
This is a 65-year-old male who had a right TKA performed outside the VA in November 2015. The patient has a history of diabetes mellitus, hypertension, lumbar discitis, hyperlipidemia, bipolar II, right bundle branch block, and chronic kidney disease. Additionally, he had a transmetatarsal amputation of the foot in late 2015.

In June 2016, he was diagnosed with L2-L3 osteomyelitis-discitis and treated with several irrigations and debridements to his spine in addition to IV antibiotics. He also admitted to leaving AMA on multiple occasions during IV antibiotic therapy. In early November 2016, he had lumbar hardware removal with subsequent wound closure on November 28, 2016 at an outside facility. He was treated with vancomycin and cefepime as his initial wound cultures revealed Candida. MRI of the lumbar spine dated December 9, 2016, revealed changes of discitis at L2-L3 with anterior vertebral wedging and kyphotic deformity, with moderate central canal stenosis at L2-L3.

On January 26, 2017 orthopedics was consulted for right knee erythema, edema, and pain. Arthrocentesis was performed on January 27, 2017 and cultures showed methicillin resistant Staphylococcus aureus (MRSA) with sensitivity to vancomycin, tigecycline, and linezolid. On January 31, 2017, right TKA removal, I&D, and insertion of antibiotic spacer was performed.  The spacer used was Cobalt G-Hv™ bone cement loaded with two grams vancomycin per bag of cement.

From January 26, 2017 to February 5, 2017 the patient was on IV vancomycin. IV antibiotics were subsequently changed to daptomycin 375 mg IV Q24Hr on February 6, 2017 due to a rising serum creatinine. MRI of the lumbar spine performed on February 26, 2017 showed residual changes of L2-L3 osteomyelitis-discitis compared to the study performed on January 27, 2017. On March, 14 2017, IV daptomycin was discontinued following a six-week course.  The patient is currently awaiting spine clearance prior to proceeding with possible re-implantation.

Patel graphs 2

The first patient’s history of IVDA may suggest that a bacteremia or superimposed subclinical endocarditis might have been the source of infection. Incomplete treatment of a previous infection may have led to seeding of the lumbar intervertebral disks as well as seeding of the prosthesis in both knees, which would serve as a reservoir for bacterial growth and biofilm development.

The first patient’s bilateral prosthetic knee infections were caused by MSSA which was also the cause of his bacteremia and is suspected in his lumbar discitis-osteomyelitis.  Prior evidence suggests that S.aureus is the predominant organism in hematogenous causes of PJI8. In a previous prospective study of 53 patients with prosthetic joints and 27 patients with other orthopedic prosthetic devices, patients were evaluated for the risk of prosthesis infection after S.aureus bacteremia. It was determined that the risk of a prosthetic joint becoming infected by hematogenous seeding of S.aureus was 34% and 7% for other orthopedic prosthetic devices9.  Another study suggests that the rate of PJI after S.aureus bacteremia ranges from 30% to 40%10.

The second case demonstrates PJI caused by MRSA in a patient with previous osteomyelitis-discitis in a non-IVDA. This lends evidence to the idea that both cases were caused by hematogenous seeding from the lumbar spine. Additionally, the second case involved a patient with diabetes mellitus. Prior studies indicate that patients with diabetes are at a significantly higher risk of developing PJI throughout their lifetimes11.

Based on a retrospective record review performed by Pigrau et al on all cases of vertebral osteomyelitis from 1986 to 2002 in one hospital it was determined that 4.6% of patients with infectious endocarditis also had pyogenic vertebral osteomyelitis. Among patients with pyogenic vertebral osteomyelitis 30.8% had infectious endocarditis12. This leads us to believe that the patient’s initial source of infection was likely the discitis-osteomyelitis causing hematogenous spread to endocardial tissue as well as to both knee prostheses.

In treatment of PJI by S.aureus one study demonstrates debridement, antibiotics, irrigation, and retention (DAIR) resulted in 55% implant retention. They also speculate that the use of rifampin may allow for equalization of prognosis for MRSA and MSSA caused PJI13.

Currently, two stage revision is the gold standard for chronically infected total knee arthroplasties. The average time between first stage resection and reimplantation is 4.4 months. This results in approximately 95.6% success rate in eradication of infection when using a cement spacer loaded with 0.55g of gentamicin in addition to empirically double IV antibiotics until sensitivities are obtained. IV antibiotics are continued during inpatient stay and are switched to oral antibiotics for five weeks following discharge. ESR and CRP are used to monitor response rate along with clinical signs of infection14. In our first case, the spacer was not removed for nearly 23 months before reimplantation was performed on the right knee. The left knee has not yet been reimplanted. The question this poses is how are outcomes for patients with long intervals before reimplantation.

In our first case, it is difficult to assess where the infection initially took root. As an IV drug abuser he is at a higher risk for infection, including bacteremia and endocarditis. Endocarditis and bacteremia could have caused hematogenous spread to his lumbar vertebrae and to both prosthetic knee joints. The likelihood that he had spontaneous latent infection to cause late onset PJI is unlikely considering the other potential sources of infection. Finally, it is also worth considering latent discitis-osteomyelitis of the lumbar vertebrae as a source of bacteremia progressing to endocarditis and eventual bilateral PJI.

One way to prevent such cases from occurring may be through increased preoperative screening. Since IV drug abusers are at increased risk of infection in general they should be excluded from participating in total joint arthroplasty unless they screen negative for a certain period of time. Another option is to increase awareness of the risks to IV drug abusers. This can be extended to other patients with significant comorbidities for prosthesis failure.

These two cases suggest the need for increased duration of antibiotic coverage in prosthetic joint patients presenting with osteomyelitis-discitis. A comprehensive protocol is necessary to better monitor these patients for PJI by identifying the infection at an earlier stage. These patients may need monthly follow-up until MRI demonstrates resolution of the osteomyelitis-discitis. Additionally, two-stage revision is the best option for these patients and it may be necessary to lengthen the spacer interval between wound sterility and reimplantation.


  1. Cochran AR, Ong KL, Lau E, Mont MA, Malkani AL. Risk of Reinfection After Treatment of Infected Total Knee Arthroplasty. J Arthroplasty. 2016;31(9 Suppl):156-161.
  2. Ruiz D, Jr., Koenig L, Dall TM, et al. The direct and indirect costs to society of treatment for end-stage knee osteoarthritis. J Bone Joint Surg Am. 2013;95(16):1473-1480.
  3. Elie Berbari MD LBM. Prosthetic joint infection: Treatment. Accessed 3/7/17.
  4. Della Valle C, Parvizi J, Bauer TW, et al. American Academy of Orthopaedic Surgeons Clinical Practice Guideline on: The Diagnosis of Periprosthetic Joint Infections of the Hip and Knee. JBJS. 2011;93(14):1355-1357.
  5. Zmistowski B, Tetreault MW, Alijanipour P, Chen AF, Della Valle CJ, Parvizi J. Recurrent periprosthetic joint infection: persistent or new infection? J Arthroplasty. 2013;28(9):1486-1489.
  6. Kamath AF, Ong KL, Lau E, et al. Quantifying the Burden of Revision Total Joint Arthroplasty for Periprosthetic Infection. The Journal of Arthroplasty.30(9):1492-1497.
  7. Shahi A, Tan TL, Chen AF, Maltenfort MG, Parvizi J. In-Hospital Mortality in Patients With Periprosthetic Joint Infection. J Arthroplasty. 2017;32(3):948-952.e941.
  8. Deacon JM, Pagliaro AJ, Zelicof SB, Horowitz HW. Prophylactic use of antibiotics for procedures after total joint replacement. J Bone Joint Surg Am. 1996;78(11):1755-1770.
  9. Murdoch DR, Roberts SA, Fowler JVG, et al. Infection of Orthopedic Prostheses after Staphylococcus aureus Bacteremia. Clin Infect Dis. 2001;32(4):647-649.
  10. Sendi P, Banderet F, Graber P, Zimmerli W. Periprosthetic joint infection following Staphylococcus aureus bacteremia. J Infect. 2011;63(1):17-22.
  11. Kunutsor SK, Whitehouse MR, Blom AW, Beswick AD. Patient-Related Risk Factors for Periprosthetic Joint Infection after Total Joint Arthroplasty: A Systematic Review and Meta-Analysis. PLoS One. 2016;11(3):e0150866.
  12. Pigrau C, Almirante B, Flores X, et al. Spontaneous pyogenic vertebral osteomyelitis and endocarditis: Incidence, risk factors, and outcome. The American Journal of Medicine. 2005;118(11):1287.e1217-1287.e1224.
  13. Lora-Tamayo J, Murillo O, Iribarren JA, et al. A Large Multicenter Study of Methicillin–Susceptible and Methicillin–Resistant Staphylococcus aureus Prosthetic Joint Infections Managed With Implant Retention. Clin Infect Dis. 2013;56(2):182-194.
  14. Silvestre A, Almeida F, Renovell P, Morante E, López R. Revision of Infected Total Knee Arthroplasty: Two-Stage Reimplantation Using an Antibiotic-Impregnated Static Spacer. Clin Orthop Surg. 2013;5(3):180-187.

Back to top