Advertisement

Predictors of Success With Chronic Antibiotic Suppression for Prosthetic Joint Infections

Published:February 07, 2022DOI:https://doi.org/10.1016/j.arth.2022.02.003

      Abstract

      Background

      Management of recurrent prosthetic joint infection (PJI) after attempted surgical eradication remains a challenge. Chronic antibiotic suppression (CAS) is regarded as a reasonable treatment option for select patients with persistent infection or multiple comorbidities. The study seeks to compare cohorts who succeed and fail with CAS.

      Methods

      This retrospective cohort study assesses patients who were treated with CAS for a PJI. Patients were included if they had a culture-proven PJI and received chronic suppressive antibiotics. Failure of suppression was defined as reoperation after initiating CAS or death occurring as result of infection. A Cox proportional hazards multivariate regression model was used to estimate risk of reoperation as a function of risk factors related to patient comorbidities, surgical history, affected joint, and infecting organism.

      Results

      We identified 45 PJIs (31 knees, 14 hips) managed with CAS with a median follow-up of 50 (95% confidence interval [CI] 33.61-74.02) months. The overall success rate of managing PJI with CAS was 67% (30/45). Controlling for body mass index and Gram status of the organism, total hip arthroplasty patients were less likely than total knee arthroplasty patients to require reoperation (hazard ratio 0.18, 95% CI 0.01-0.96, P = .04). Patients with Gram-positive infections were less likely than those with a Gram-negative infections to require reoperation (hazard ratio 0.22, 95% CI 0.05-0.88, P = .03). Severe antibiotic side effects were rare. Patients who experienced multiple changes to their antibiotic regimen were more likely to fail with CAS.

      Conclusion

      CAS is a reasonable strategy in patients with PJI who lack or refuse further surgical treatment options. Most hips and Gram-positive infections treated with CAS successfully avoided reoperation in this cohort.

      Keywords

      Prosthetic joint infection (PJI) is a devastating and severe complication of total joint arthroplasty resulting in mortality, morbidity, and financial burden for the patient and healthcare system [
      • Berend K.R.
      • Lombardi A.V.
      • Morris M.J.
      • Bergeson A.G.
      • Adams J.B.
      • Sneller M.A.
      Two-stage treatment of hip periprosthetic joint infection is associated with a high rate of infection control but high mortality.
      ,
      • Kurtz S.M.
      • Lau E.
      • Watson H.
      • Schmier J.K.
      • Parvizi J.
      Economic burden of periprosthetic joint infection in the United States.
      ,
      • Zmistowski B.
      • Karam J.A.
      • Durinka J.B.
      • Casper D.S.
      • Parvizi J.
      Periprosthetic joint infection increases the risk of one-year mortality.
      ]. Although infection after index total joint arthroplasty is uncommon in 1%-2% of cases [
      • Pulido L.
      • Ghanem E.
      • Joshi A.
      • Purtill J.J.
      • Parvizi J.
      Periprosthetic joint infection: the incidence, timing, and predisposing factors.
      ], PJI is the leading cause for failure of total knee arthroplasty (TKA, 25%) and is a common indication for revision hip arthroplasty (15%) [
      • Bozic K.J.
      • Kurtz S.M.
      • Lau E.
      • Ong K.
      • Chiu V.
      • Vail T.P.
      • et al.
      The epidemiology of revision total knee arthroplasty in the United States.
      ,
      • Bozic K.J.
      • Kurtz S.M.
      • Lau E.
      • Ong K.
      • Vail T.P.
      • Berry D.J.
      The epidemiology of revision total hip arthroplasty in the United States.
      ].
      Surgical management of PJI can range from debridement, antibiotics, and implant retention, to single-stage or 2-stage revision [
      • Li C.
      • Renz N.
      • Trampuz A.
      Management of periprosthetic joint infection.
      ]. These options vary depending on infection acuity, patient comorbidities, affected joint, infecting organism, and treatment philosophy. Currently, the gold standard for eradiation of a chronic PJI consists of a 2-stage revision involving removal of components, placement of an antibiotic-loaded spacer, extended systemic antibiotics, and reimplantation. Despite optimal surgical and medical outcomes, the incidence of reinfection following 2-stage revision for PJI in the hip and knee is reported between15% and 32% [
      • Petis S.M.
      • Abdel M.P.
      • Perry K.I.
      • Mabry T.M.
      • Hanssen A.D.
      • Berry D.J.
      Long-term results of a 2-stage exchange protocol for periprosthetic joint infection following total hip arthroplasty in 164 hips.
      ,
      • Petis S.M.
      • Perry K.I.
      • Mabry T.M.
      • Hanssen A.D.
      • Berry D.J.
      • Abdel M.P.
      Two-stage exchange protocol for periprosthetic joint infection following total knee arthroplasty in 245 knees without prior treatment for infection.
      ,
      • Kunutsor S.K.
      • Whitehouse M.R.
      • Blom A.W.
      • Beswick A.D.
      Re-infection outcomes following one- and two-stage surgical revision of infected hip prosthesis: a systematic review and meta-analysis.
      ,
      • Kunutsor S.K.
      • Whitehouse M.R.
      • Lenguerrand E.
      • Blom A.W.
      • Beswick A.D.
      Re-infection outcomes following one- and two-stage surgical revision of infected knee prosthesis: a systematic review and meta-analysis.
      ,
      • Kunutsor S.K.
      • Whitehouse M.R.
      • Blom A.W.
      • Board T.
      • Kay P.
      • Wroblewski B.M.
      • et al.
      One- and two-stage surgical revision of peri-prosthetic joint infection of the hip: a pooled individual participant data analysis of 44 cohort studies.
      ]. When reinfection occurs after 2-stage revision, reoperation with an additional 2-stage treatment may be attempted. Unfortunately, while subsequent reoperation provides another opportunity to cure the infection, success rates are often lower due to diminished bone and soft tissue quality [
      • Azzam K.
      • McHale K.
      • Austin M.
      • Purtill J.J.
      • Parvizi J.
      Outcome of a second two-stage reimplantation for periprosthetic knee infection.
      ,
      • Mortazavi S.M.
      • O'Neil J.T.
      • Zmistowski B.
      • Parvizi J.
      • Purtill J.J.
      Repeat 2-stage exchange for infected total hip arthroplasty: a viable option?.
      ]. Additionally, there are a portion of patients with reinfection who refuse additional surgery, who are not optimal surgical candidates due to medical comorbidities, or who would require limb-threatening surgery.
      Chronic suppression of PJI with antibiotics is a reasonable alternative treatment option for this subset of patients who refuse or cannot undergo additional surgery [
      • Wouthuyzen-Bakker M.
      • Nijman J.
      • Kampinga G.
      • Assen S.
      • Jutte P.
      Efficacy of antibiotic suppressive therapy in patients with a prosthetic joint infection.
      ,
      • Siqueira M.B.
      • Saleh A.
      • Klika A.K.
      • O'Rourke C.
      • Schmitt S.
      • Higuera C.A.
      • et al.
      Chronic suppression of periprosthetic joint infections with oral antibiotics increases infection-free survivorship.
      ,
      • Pavoni G.L.
      • Giannella M.
      • Falcone M.
      • Scorzolini L.
      • Liberatore M.
      • Carlesimo B.
      • et al.
      Conservative medical therapy of prosthetic joint infections: retrospective analysis of an 8-year experience.
      ]. The goal of chronic antibiotic suppression (CAS) therapy in this context is to control clinical symptoms rather than to cure infection. There is limited existing literature on outcomes or predictors of success with CAS therapy [
      • Malahias M.A.
      • Gu A.
      • Harris E.C.
      • Adriani M.
      • Miller A.O.
      • Westrich G.H.
      • et al.
      The role of long-term antibiotic suppression in the management of peri-prosthetic joint infections treated with debridement, antibiotics, and implant retention: a systematic review.
      ]. Therefore, the purpose of this study is to describe the clinical outcomes in a cohort of patients who received CAS for a PJI after TKA or total hip arthroplasty (THA). A second aim is to identify factors that would predict success with suppression in order to guide future management. Finally, this study aims to characterize complications related to antibiotics and how they impact the treatment plan.

      Materials and Methods

      Once approved by the Institutional Review Board, patient cases were identified using relevant CPT codes (27447, 27486, 27130, 27132, 27134) and key terms (PJI, CAS) followed by manual review of the electronic medical record. Patients aged >18 years with culture-proven PJI who were started on antibiotics with documented intent for chronic suppression between 2007 and 2020 were eligible for inclusion in the study. We identified 48 total PJIs treated with CAS in 45 patients. Among this group, there were 3 patients who had 2 PJIs, all with 1 knee and 1 hip successfully treated with CAS at last known follow-up. Because the event of repeat humans was rare and required a more complicated survival model unlikely to impact study outcomes, only the first joint diagnosed for these patients was included in the final analysis (2 hips, 1 knee). Therefore, there were 45 PJIs in 45 unique patients included in the final analysis.
      Demographic data, comorbidities, ambulatory status, the date of PJI diagnosis, and number, dates, and types of revision surgeries were recorded. The primary infecting organism and antibiotic course was also observed. The start date of CAS therapy was determined through review of infectious disease and orthopedic notes. The chronic antibiotic suppressive regimen, dosing, and monitoring for patients included in this study were predominantly managed by the infectious disease team. CAS was considered successful if the patient did not require reoperation after starting CAS therapy. CAS was considered a failure if the patient underwent reoperation after starting CAS therapy or if they died of causes directly related to their PJI.

      Statistical Analysis

      Univariable and multivariable Cox proportional hazards models were used to estimate the instantaneous risk of reoperation at any given time following the index infection date as a function of patient age, gender, race, body mass index (BMI), number of prior surgeries, type of spacer, surgery type, infection-free duration, Charlson comorbidity score, ambulatory status, and infecting organism. On univariable analysis, confidence limits and significance values (P-values) for the spacer type and infecting organism were adjusted for multiple comparisons using a Sidak correction, and ambulatory status was treated as a quantitative ordinal. Covariates included in the multivariable model were those maximized the model’s score statistic and improved the model’s fit statistic (as measured by Akaike’s information criterion or AIC).
      In these time-to-event models, elapsed time was measured in months from the index infection date to date of reoperation, and patients not experiencing reoperation were censored at their last known follow-up date. The proportional hazard assumption for each covariate was assessed graphically using Martingale residuals [
      • Lin D.Y.
      • Wei L.J.
      • Ying Z.
      Checking the Cox model with cumulative sums of martingale-based residuals.
      ]. Finally, traditional Kaplan-Meier curves were used to estimate the median event-free time distribution among all patients which was further stratified by surgery type. Median follow-up time was assessed using a reverse Kaplan-Meier curve [
      • Schemper M.
      • Smith T.L.
      A note on quantifying follow-up in studies of failure time.
      ]. Antibiotic side effects and changes are reported descriptively.

      Results

      We identified 45 PJIs (31 knees, 14 hips) in 45 unique individuals managed with CAS between 2007 and 2020 with a median follow-up of 50 months (95% confidence interval [CI] 33.61-74.02). The overall success rate in avoiding reoperation after CAS in this cohort was 67% (30/45). Success rate in TKAs was 55% (17/31) and success rate in THAs was 93% (13/14). Available data showed no mortalities directly related to PJI in this cohort, though 3 deaths did occur due to causes felt to be unrelated to CAS treatment (1 pneumonia, 1 myocardial infarction, 1 acute cardiorenal syndrome). PJI was diagnosed at a median of 24 (interquartile range [IQR] 2-64) months following the index surgery in the group that was successfully treated with CAS, and at 33 (IQR 2-43) months in the group that failed management with CAS. Patients with successful management with CAS underwent a median of 2 (IQR 1-3) subsequent surgeries for infection prior to starting CAS, while patients with failed management with CAS underwent a median of 3 (IQR 2-4) subsequent surgeries. Staphylococcus was the most common primary infecting organism, comprising 62% of the total infections. Gram-negative infections comprised 13% (4/30) of the TKAs, and 6% (1/15) of the THAs in the cohort, and the remainder were Gram positive. Of the patients successfully managed with CAS, 26% (8/30) had sinus tracts at the last known follow-up. Summary frequencies of the demographic data, prior surgical history, and details regarding the infecting organism are shown in Table 1. Oral antibiotic regimens for the patients included in the study are listed in Table 2.
      Table 1Summary Frequencies.
      VariableReoperationTotal (n = 45)
      No (n = 30)Yes (n = 15)
      Female17 (56.7%)9 (60.0%)26 (57.8%)
      Median age (IQR)63 (58-72)61 (54-76)62 (57-73)
      Median BMI (IQR)31 (24-38)35 (31-43)33 (27-40)
      Race
       White25 (83.3%)13 (86.7%)38 (84.4%)
       Black3 (10.0%)2 (13.3%)5 (11.1%)
       Other2 (6.7%)0 (0.0%)2 (4.4%)
      Ethnicity
       Non-Hispanic28 (96.6%)14 (93.3%)42 (95.5%)
       Hispanic1 (3.4%)1 (6.7%)2 (4.5%)
      Median CCI (IQR)4 (2-5)4 (1-7)4 (2-5)
      Index surgery
       TKA17 (56.7%)14 (93.3%)31 (68.9%)
       THA13 (43.3%)1 (6.7%)14 (31.1%)
      Median number of surgeries after index prior to starting CAS (IQR)2 (1-3)3 (1-4)3 (1-3)
      Spacer type used in revision surgery
       Static10 (33.3%)5 (33.3%)15 (33.3%)
       Articulating8 (26.7%)7 (46.7%)15 (33.3%)
       None12 (40.0%)3 (20.0%)15 (33.3%)
      Ambulatory status at the start of CAS
       Independent25 (83.3%)12 (80.0%)37 (82.2%)
       Cane3 (10.0%)0 (0.0%)3 (6.7%)
       Walker1 (3.3%)3 (20.0%)4 (8.9%)
       Wheelchair/bedbound1 (3.3%)0 (0.0%)1 (2.2%)
      Staphylococcus status
       Negative9 (30.0%)8 (53.3%)17 (37.8%)
       Positive21 (70.0%)7 (46.7%)28 (62.2%)
      Methicillin-sensitive Staphylococcus9 (30.0%)3 (20.0%)12 (26.7%)
      Methicillin-resistant Staphylococcus4 (13.3%)4 (26.7%)8 (17.8%)
      Coagulase-negative Staphylococcus8 (26.7%)0 (0.0%)8 (17.8%)
      Gram status
       Positive28 (93.3%)12 (80.0%)40 (88.9%)
       Negative2 (6.7%)3 (20.0%)5 (11.1%)
      IQR, interquartile range; BMI, body mass index; TKA, total knee arthroplasty; THA, total hip arthroplasty; CCI, Charlson Comorbidity Index; CAS, chronic antibiotic suppression; MRSA, methicillin-resistant Staphylococcus aureus.
      Table 2Antibiotic Summary.
      PatientAffected JointMicroorganismOutcomeOral Suppressive Regimen 1Change of TreatmentOral Suppressive Regimen 2Change of Treatment
      1KneeMSSASuccessCephalexin + FluconazoleYes. Fluconazole stopped for hallucinationsCephalexinNo
      2HipMSSASuccessCephalexin + RifampinYes. Rifampin stopped for failure to control symptomsCephalexinNo
      3HipCandida albicansSuccessFluconazole + AmoxicillinNo
      4KneeMSSASuccessClindamycinNo
      5KneeMSSASuccessTrimethoprim/sulfamethoxazole + RifampinNo
      6KneeCoagulase-negative StaphylococcusSuccessLinezolidNo. Tooth discoloration noted, not stopped
      7KneeMRSASuccessClindamycinNo
      8HipCoagulase-negative StaphylococcusSuccessCephalexinNo
      9HipMSSASuccessCephalexinNo
      10HipCoagulase-negative StaphylococcusSuccessTrimethoprim/sulfamethoxazole + RifampinYes. Rifampin stopped, drug interaction with anticoagulantTrimethoprim/sulfamethoxazoleNo
      11KneeCoagulase-negative StaphylococcusSuccessDoxycyclineNo
      12HipMSSASuccessLevofloxacinNo
      13HipPolymicrobialSuccessAugmentinNo
      14KneeStreptococcus agalactiaeSuccessCephalexinNo
      15HipS agalactiaeSuccessMoxifloxacinNo
      16KneeMRSASuccessDoxycyclineYes. Stopped due to failure of treatmentCephalexinNo
      17KneeCoagulase-negative StaphylococcusSuccessCiprofloxacinYes. Stopped due to dosing scheduleLevofloxacinNo
      18KneePseudomonas aeruginosaSuccessAmoxicillin/Clavulanic acidYes. Stopped due to failure of treatmentTrimethoprim/sulfamethoxazoleNo
      19KneeMRSASuccessLinezolidYes. Stopped due to renal failureDoxycyclineNo
      20KneeMSSASuccessCephalexinNo
      21KneeMSSASuccessCephalexinNo
      22HipCoagulase-negative StaphylococcusSuccessDoxycyclineNo
      23KneeStreptococcus pneumoniaeSuccessClindamycinYes. Stopped due to reason not listedPenicillin-VNo
      24KneeMSSASuccessCephalexinNo
      25HipMRSASuccessTrimethoprim/sulfamethoxazole
      26HipProteus mirabilisSuccessCephalexinNo. Additional antibiotic added for symptom managementCephalexin + DoxycyclineNo
      27KneeCoagulase-negative StaphylococcusSuccessDoxycycline + RifampinNo. Nausea noted, treatment not changed
      28KneeMSSASuccessDoxycyclineYes. Stopped due to failure of treatmentCephalexinNo
      29HipPolymicrobialSuccessDoxycyclineNo
      30HipCoagulase-negative staphylococcusSuccessDoxycyclineYes. Stopped due to costTrimethoprim/sulfamethoxazoleNo
      31KneeP mirabilisFailureDoxycyclineYes. Stopped due to failure of treatment
      32KneeStreptococcus viridansFailureLevaquinYes. Stopped; reason not listedAmoxicillinYes. Stopped due to failure of treatment
      33KneePropionibacterium acnesFailureAmoxicillinYes. Stopped due to failure of treatment
      34KneeMSSAFailureDoxycyclineYes. Stopped due to failure of treatment
      35KneeEnterococcusFailureCephalexinYes. Stopped due to failure of treatment
      36KneeMRSAFailureAmoxicillin/Clavulanic acidYes. Stopped due to failure of treatmentDoxycyclineYes. Stopped due to failure of treatment
      37KneeMSSAFailureCephalexinYes. Stopped due to failure of treatment
      38KneePasteurella multiformeFailureTrimethoprim/sulfamethoxazoleYes. Stopped due to failure of treatment
      39HipPolymicrobialFailureDoxycycline + CiprofloxacinYes. Stopped due to failure of treatment
      40KneeMRSAFailureTrimethoprim/sulfamethoxazoleYes. Stopped due to failure of treatment
      41KneeMSSAFailureCefadroxilYes. Stopped due to failure of treatmentTrimethoprim/sulfamethoxazoleYes. Stopped due to failure of treatment
      42KneeMRSAFailureTrimethoprim/sulfamethoxazoleYes. Stopped due to failure of treatmentLinezolidYes. Stopped due to failure of treatment. Tooth discoloration also noted
      43KneeCorynebacteriumFailureDoxycyclineYes. Stopped due to failure of treatment
      44KneeMRSAFailureAmoxicillinYes. Stopped due to failure of treatmentTrimethoprim/sulfamethoxazoleYes. Stopped due to failure of treatment
      45KneeS agalactiaeFailureTrimethoprim/sulfamethoxazoleYes. Stopped due to renal failureDoxycyclineYes. Stopped due to diarrhea
      MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus.
      Controlling for patient BMI and the Gram status of the infecting organism, THA patients were less likely than TKA patients to require reoperation during the follow-up period (hazard ratio [HR] 0.18, 95% CI 0.01-0.96, P = .04). Controlling for all other variables in the model, patients with a Gram-positive infection were less likely than patients with a Gram-negative infection to require reoperation (HR 0.22, 95% CI 0.05-0.88, P = .03). There were no statistically significant differences in reoperation risk after starting CAS in regards to age, gender, CCI, BMI, and number or type of revision surgeries. Reoperation risk as it relates to variables of interest is summarized in Table 3.
      Table 3Risk of Reoperation.
      VariableUnadjustedAdjusted
      Hazard Ratio (95% CI)P ValueHazard Ratio (95% CI)P Value
      Age (per 5-y increase)1.04 (0.83-1.30).72
      Gender: male vs female0.92 (0.32-2.58).87
      Race: non-White vs White0.61 (0.13-2.78).52
      BMI (per 5 kg/m2 increase)1.19 (0.93-1.53).161.15 (0.88-1.50).31
      Prior surgeries (per 1 count increase)1.03 (0.74-1.42).88
      Spacer type.74
      Overall type III test.
       Articulating vs static1.39 (0.34-5.71).58
       Articulating vs none1.63 (0.31-8.74).48
       Static vs none1.18 (0.20-6.78).82
      Surgery type: THA vs TKA0.14 (0.01-0.72).010.18 (0.01-0.96).04
      Infection-free duration (per year increase)0.99 (0.88-1.12).90
      CCI (per 1-point increase)1.07 (0.84-1.36).58
      Ambulatory status (per 1-stage increase)1.05 (0.56-2.00).87
      Infection type
       Staphylococcus (yes vs no)0.37 (0.10-1.37).21
       MRSA (yes vs no)1.20 (0.28-5.17).99
       Gram status (positive vs negative)0.21 (0.04-1.21).100.22 (0.05-0.88).03
      Valid N = 45 with 15 (33%) events. Confidence limits and significance values (P-values) for the type of spacer and type of infection are adjusted for multiple comparisons using a Sidak correction.
      BMI, body mass index; CCI, Charlson Comorbidity Index; THA, total hip arthroplasty; TKA, total knee arthroplasty; CI, confidence interval; MRSA, methicillin-resistant Staphylococcus aureus.
      a Overall type III test.
      It takes 61.3 months for the event-free probability of reoperation to drop below 50% for those receiving CAS for PJI following TKA. Conversely, the event-free probability of reoperation never drops below 90% for those receiving CAS for PJI following THA (Fig. 1). Likewise, it takes only 28.5 months for the event-free probability of reoperation to drop below 50% for patients with Gram-negative PJIs. Conversely, it takes 116.6 months for the event-free probability of reoperation to drop below 50% for patients with Gram-positive PJIs (Fig. 2).
      Figure thumbnail gr1
      Fig. 1Cumulative event-free estimates for surgery type. THA, total hip arthroplasty; TKA, total knee arthroplasty.
      Figure thumbnail gr2
      Fig. 2Cumulative event-free estimates for Gram status of infecting organism.
      Patients who failed management with CAS were more likely to have their antibiotic regimen changed during their treatment course. Among patients who failed CAS, 7% (1/15) had no changes to their regimen, 60% (9/15) had 1 change, 20% (3/15) had 2 changes, and 13% (2/15) had 3 or more changes. Of patients successfully managed with CAS, 63% (19/30) had no change in their antibiotic regimen and 37% (11/30) had 1 change. The most common reason for changing an antibiotic was a recurrence of symptoms in the affected joint, which occurred 4 times in the successful treatment group and 14 times in the failed treatment group. Side effects that resulted in discontinuing an antibiotic were acute kidney injury (2), neurotoxicity (1), hallucinations (1), diarrhea (1), and tooth discoloration (1). Side effects that did not result in a change in discontinuing an antibiotic were vaginal thrush (1), nausea (2), diarrhea (2), and tooth discoloration (1). Antibiotics were also changed due to cost (1) and dosing schedule (1).

      Discussion

      In this retrospective cohort study, we present a 67% (30/45) success rate for managing PJI with CAS, where success is defined as avoiding reoperation on the affected joint and avoiding mortality due to infection. CAS was more likely to be successful in the setting of THA or Gram-positive PJI. No other factors assessed predicted a successful outcome as defined by the study. To our knowledge, this is the largest retrospective review assessing outcomes with CAS for PJI. Our results indicate that CAS provides moderate success for managing chronic infection in patients with PJIs who are not candidates for further surgical treatment.
      Our findings agree with the results of prior studies reviewing CAS outcomes with success rates in the range of 63%-86% [
      • Wouthuyzen-Bakker M.
      • Nijman J.
      • Kampinga G.
      • Assen S.
      • Jutte P.
      Efficacy of antibiotic suppressive therapy in patients with a prosthetic joint infection.
      ,
      • Siqueira M.B.
      • Saleh A.
      • Klika A.K.
      • O'Rourke C.
      • Schmitt S.
      • Higuera C.A.
      • et al.
      Chronic suppression of periprosthetic joint infections with oral antibiotics increases infection-free survivorship.
      ,
      • Pavoni G.L.
      • Giannella M.
      • Falcone M.
      • Scorzolini L.
      • Liberatore M.
      • Carlesimo B.
      • et al.
      Conservative medical therapy of prosthetic joint infections: retrospective analysis of an 8-year experience.
      ]. A more recent study by Wouthuyzen-Bakker et al [
      • Wouthuyzen-Bakker M.
      • Nijman J.
      • Kampinga G.
      • Assen S.
      • Jutte P.
      Efficacy of antibiotic suppressive therapy in patients with a prosthetic joint infection.
      ] closely matches our results with a 67% treatment success for CAS in their cohort of 21 patients. Furthermore, this group reported higher success rates with CAS in PJIs in which Staphylococcus was the primary infecting organism [
      • Wouthuyzen-Bakker M.
      • Nijman J.
      • Kampinga G.
      • Assen S.
      • Jutte P.
      Efficacy of antibiotic suppressive therapy in patients with a prosthetic joint infection.
      ]. Traditionally, Gram-negative organisms are considered to be more difficult to treat with lower success rates with surgical management of PJI [
      • Hsieh P.-H.
      • Lee M.S.
      • Hsu K.-Y.
      • Chang Y.H.
      • Shih H.N.
      • Ueng S.W.
      • et al.
      Gram-negative prosthetic joint infections: risk factors and outcome of treatment.
      ,
      • Zmistowski B.
      • Fedorka C.J.
      • Sheehan E.
      • Deirmengian G.
      • Austin M.S.
      • Parvizi J.
      Prosthetic joint infection caused by gram-negative organisms.
      ,
      • da Silva R.B.
      • Salles M.J.
      Outcomes and risk factors in prosthetic joint infections by multidrug-resistant gram-negative bacteria: a retrospective cohort study.
      ]. The lower success rates may be due to additional biofilm mechanisms of these organisms, which can impact both antibiotic penetration and successful surgical debridement [
      • da Silva R.B.
      • Salles M.J.
      Outcomes and risk factors in prosthetic joint infections by multidrug-resistant gram-negative bacteria: a retrospective cohort study.
      ]. The greater success with CAS of PJIs in hips in this study is unclear and not commented elsewhere in available literature. A possible mechanism for this may be related to the larger soft tissue envelope around the hip and thus decreased incidence of the same wound healing complications that the knee is subject to.
      Chronic antibiotic use warrants close partnership with infectious disease specialists to select culture-driven regimens and to monitor for clinical and laboratory evidence of failed suppression. Although our cohort is too small to draw statistical conclusions, multiple changes to the antibiotic regimen appear to be a poor prognostic indicator. Overall, antibiotics were well tolerated with a low number of side effects. Severe side effects from antibiotics such as renal toxicity or neurotoxicity were rare in this cohort.
      This study has several limitations that are worth noting. The sample size and number of reoperation events are low and result in wide confidence limits. However, the results from the regression analysis agree with the small sample univariate analysis which supports that these are true results rather than statistical chance findings. Additionally, the sample size in this study is the largest available in the current literature. The conclusion for the Gram status as a predictor of successful treatment should be replicated in a larger prospective study of patients at-risk for reoperation, and the current work can partially inform the sample size calculation for that study. The study findings are also limited by the retrospective nature of the study and the relatively limited duration of follow-up. A higher mortality rate was expected in this population than was observed. This may be accounted for in missing data in those patients lost to follow-up. Because pain and soft tissue findings are reported inconsistently in the available charts, standard PJI success definitions could not be used here [
      • Tan T.L.
      • Goswami K.
      • Fillingham Y.A.
      • Shohat N.
      • Rondon A.J.
      • Parvizi J.
      Defining treatment success after 2-stage exchange arthroplasty for periprosthetic joint infection.
      ]. Finally, patient-reported outcomes are not available, and success defined as avoiding reoperation may oversimplify this challenging clinical scenario and not match patients’ personal definitions of successful treatment.

      Conclusions

      In conclusion, CAS is a reasonable strategy in patients with PJI who lack, defer, or are unable to undergo further surgical treatment options. Most hips and Gram-positive infections treated with CAS in this cohort were able to successfully avoid reoperation or infection-related mortality related to their infection. Time to event analysis for various patient and infection scenarios can enable the treating surgeon to provide improved guidance and expectations to patients with this challenging clinical problem.

      Appendix A. Supplementary Data

      References

        • Berend K.R.
        • Lombardi A.V.
        • Morris M.J.
        • Bergeson A.G.
        • Adams J.B.
        • Sneller M.A.
        Two-stage treatment of hip periprosthetic joint infection is associated with a high rate of infection control but high mortality.
        Clin Orthop Relat Res. 2013; 471: 510-518https://doi.org/10.1007/s11999-012-2595-x
        • Kurtz S.M.
        • Lau E.
        • Watson H.
        • Schmier J.K.
        • Parvizi J.
        Economic burden of periprosthetic joint infection in the United States.
        J Arthroplasty. 2012; 27: 61-65.e1https://doi.org/10.1016/j.arth.2012.02.022
        • Zmistowski B.
        • Karam J.A.
        • Durinka J.B.
        • Casper D.S.
        • Parvizi J.
        Periprosthetic joint infection increases the risk of one-year mortality.
        J Bone Joint Surg Am. 2013; 95: 2177-2184https://doi.org/10.2106/JBJS.L.00789
        • Pulido L.
        • Ghanem E.
        • Joshi A.
        • Purtill J.J.
        • Parvizi J.
        Periprosthetic joint infection: the incidence, timing, and predisposing factors.
        Clin Orthop Relat Res. 2008; 466: 1710-1715https://doi.org/10.1007/s11999-008-0209-4
        • Bozic K.J.
        • Kurtz S.M.
        • Lau E.
        • Ong K.
        • Chiu V.
        • Vail T.P.
        • et al.
        The epidemiology of revision total knee arthroplasty in the United States.
        Clin Orthop Relat Res. 2010; 468: 45-51https://doi.org/10.1007/s11999-009-0945-0
        • Bozic K.J.
        • Kurtz S.M.
        • Lau E.
        • Ong K.
        • Vail T.P.
        • Berry D.J.
        The epidemiology of revision total hip arthroplasty in the United States.
        J Bone Joint Surg Am. 2009; 91: 128-133https://doi.org/10.2106/JBJS.H.00155
        • Li C.
        • Renz N.
        • Trampuz A.
        Management of periprosthetic joint infection.
        Hip Pelvis. 2018; 30: 138-146https://doi.org/10.5371/hp.2018.30.3.138
        • Petis S.M.
        • Abdel M.P.
        • Perry K.I.
        • Mabry T.M.
        • Hanssen A.D.
        • Berry D.J.
        Long-term results of a 2-stage exchange protocol for periprosthetic joint infection following total hip arthroplasty in 164 hips.
        J Bone Joint Surg Am. 2019; 101: 74-84https://doi.org/10.2106/JBJS.17.01103
        • Petis S.M.
        • Perry K.I.
        • Mabry T.M.
        • Hanssen A.D.
        • Berry D.J.
        • Abdel M.P.
        Two-stage exchange protocol for periprosthetic joint infection following total knee arthroplasty in 245 knees without prior treatment for infection.
        J Bone Joint Surg Am. 2019; 101: 239-249https://doi.org/10.2106/JBJS.18.00356
        • Kunutsor S.K.
        • Whitehouse M.R.
        • Blom A.W.
        • Beswick A.D.
        Re-infection outcomes following one- and two-stage surgical revision of infected hip prosthesis: a systematic review and meta-analysis.
        PLoS One. 2015; 10: e0139166https://doi.org/10.1371/journal.pone.0139166
        • Kunutsor S.K.
        • Whitehouse M.R.
        • Lenguerrand E.
        • Blom A.W.
        • Beswick A.D.
        Re-infection outcomes following one- and two-stage surgical revision of infected knee prosthesis: a systematic review and meta-analysis.
        PLoS One. 2016; 11: e0151537https://doi.org/10.1371/journal.pone.0151537
        • Kunutsor S.K.
        • Whitehouse M.R.
        • Blom A.W.
        • Board T.
        • Kay P.
        • Wroblewski B.M.
        • et al.
        One- and two-stage surgical revision of peri-prosthetic joint infection of the hip: a pooled individual participant data analysis of 44 cohort studies.
        Eur J Epidemiol. 2018; 33: 933-946https://doi.org/10.1007/s10654-018-0377-9
        • Azzam K.
        • McHale K.
        • Austin M.
        • Purtill J.J.
        • Parvizi J.
        Outcome of a second two-stage reimplantation for periprosthetic knee infection.
        Clin Orthop Relat Res. 2009; 467: 1706-1714https://doi.org/10.1007/s11999-009-0739-4
        • Mortazavi S.M.
        • O'Neil J.T.
        • Zmistowski B.
        • Parvizi J.
        • Purtill J.J.
        Repeat 2-stage exchange for infected total hip arthroplasty: a viable option?.
        J Arthroplasty. 2012; 27: 923-926.e1https://doi.org/10.1016/j.arth.2011.11.006
        • Wouthuyzen-Bakker M.
        • Nijman J.
        • Kampinga G.
        • Assen S.
        • Jutte P.
        Efficacy of antibiotic suppressive therapy in patients with a prosthetic joint infection.
        J Bone Joint Infect. 2017; 2: 77-83https://doi.org/10.7150/jbji.17353
        • Siqueira M.B.
        • Saleh A.
        • Klika A.K.
        • O'Rourke C.
        • Schmitt S.
        • Higuera C.A.
        • et al.
        Chronic suppression of periprosthetic joint infections with oral antibiotics increases infection-free survivorship.
        J Bone Joint Surg Am. 2015; 97: 1220-1232https://doi.org/10.2106/JBJS.N.00999
        • Pavoni G.L.
        • Giannella M.
        • Falcone M.
        • Scorzolini L.
        • Liberatore M.
        • Carlesimo B.
        • et al.
        Conservative medical therapy of prosthetic joint infections: retrospective analysis of an 8-year experience.
        Clin Microbiol Infect. 2004; 10: 831-837https://doi.org/10.1111/j.1469-0691.2004.00928.x
        • Malahias M.A.
        • Gu A.
        • Harris E.C.
        • Adriani M.
        • Miller A.O.
        • Westrich G.H.
        • et al.
        The role of long-term antibiotic suppression in the management of peri-prosthetic joint infections treated with debridement, antibiotics, and implant retention: a systematic review.
        J Arthroplasty. 2020; 35: 1154-1160https://doi.org/10.1016/j.arth.2019.11.026
        • Lin D.Y.
        • Wei L.J.
        • Ying Z.
        Checking the Cox model with cumulative sums of martingale-based residuals.
        Biometrika. 1993; 80: 557-572
        • Schemper M.
        • Smith T.L.
        A note on quantifying follow-up in studies of failure time.
        Control Clin Trials. 1996; 17: 343-346
        • Hsieh P.-H.
        • Lee M.S.
        • Hsu K.-Y.
        • Chang Y.H.
        • Shih H.N.
        • Ueng S.W.
        • et al.
        Gram-negative prosthetic joint infections: risk factors and outcome of treatment.
        Clin Infect Dis. 2009; 49: 1036
        • Zmistowski B.
        • Fedorka C.J.
        • Sheehan E.
        • Deirmengian G.
        • Austin M.S.
        • Parvizi J.
        Prosthetic joint infection caused by gram-negative organisms.
        J Arthroplasty. 2011; 26: 104-108https://doi.org/10.1016/j.arth.2011.03.044
        • da Silva R.B.
        • Salles M.J.
        Outcomes and risk factors in prosthetic joint infections by multidrug-resistant gram-negative bacteria: a retrospective cohort study.
        Antibiotics. 2021; 10: 340https://doi.org/10.3390/antibiotics10030340
        • Tan T.L.
        • Goswami K.
        • Fillingham Y.A.
        • Shohat N.
        • Rondon A.J.
        • Parvizi J.
        Defining treatment success after 2-stage exchange arthroplasty for periprosthetic joint infection.
        J Arthroplasty. 2018; 33: 3541-3546https://doi.org/10.1016/j.arth.2018.06.015