One-stage exchange for prosthetic joint infection of the hip

2022 American Association of Hip and Knee Surgeons Symposium: Periprosthetic Joint Infection

Periprosthetic joint infection (PJI) is the leading cause of failure in patients undergoing total joint arthroplasty. This article is a brief summary of a symposium on PJI that was presented at the annual AAHKS meeting. It will provide an overview of current technqiues in the prevention, diagnosis, and management of PJI. It will also highlight emerging technologies in this setting.

Comparison of Patient-Reported Outcomes Measures and Quality-Adjusted Life Years Following One- and Two-Stage Septic Knee Exchange

(1) Background: Periprosthetic joint infection (PJI) can be managed with one- or two-stage revision surgery protocol. Despite several studies analyzing the eradication rates between both procedures, there are no comparative studies that analyze patient-reported outcome measures (PROMs) and quality-adjusted life years (QALYs) in both treatment strategies. (2) Methods: All patients who underwent a two-stage knee revision between January 2017 to December 2018, due to a periprosthetic joint infection were included in the study. From the time interval, we selected a comparative group with the one-stage septic procedure. All patients received the following questionnaires: Oxford Knee Score, EQ-5D-5L, SSQ-8, and the SF-36. Additionally, demographic patient data were collected. The quality-adjusted life years (QALY) were calculated using the EQ-5D-5L. (3) Results: A total of 35 patients with a mean age of 67.7 years (SD = 8.9) were included in the final evaluation. The mean follow-up period was 54.5 months (SD = 5.5). There was no statistically significant difference regarding the Charlson Comorbidity Index (CCI), postoperative complications, or all evaluated questionnaires. There was no statistically significant difference in QALYs between the one- and two-stage revision. (4) Conclusion: Our study results show that the one-stage revision for PJI achieves similar PROMs compared to two-stage revision.

One-stage revision arthroplasty in a patient with ochronotic arthropathy accompanied by joint infection: A case report

BACKGROUND

Ochronotic arthropathy (OcA) is a rare disease, which is caused by the accumulation of homogentisic acid in the joint. Patients with OcA have obvious joint pain and the disease progresses rapidly, eventually resulting in disability. Arthroplasty is an efficacious treatment in patients with OcA. However, when OcA patients have joint infection, is joint replacement an option? In the present report, we performed total knee arthroplasty in a patient with OcA and knee infection under the guidance of one-stage revision theory.

CASE SUMMARY

A 64-year-old male was referred to our hospital due to severe left knee pain with limited mobility for 2 years. On physical examination, the patient was found to have dark brown pigmentation of the sclera and auricle. Laboratory test results showed elevations in C-reactive protein level (65.79 mg/L) and erythrocyte sedimentation rate (90.00 mm/h). The patient underwent debridement of the left knee joint, during which the cartilage surface of the knee joint was found to be black-brown in color. Bacterial culture of synovial fluid revealed Achromobacter xylosoxidans. We then carried out arthroplasty under the guidance of the theory of one-stage revision. After surgery, the patient's left knee joint pain disappeared and function recovered without joint infection.

CONCLUSION

OcA accompanied by joint infection is rare. One-stage revision arthroplasty may be a treatment option for this disease.

“Unexpected” Infections in Revision Arthroplasty for Aseptic Loosening

Background. Data from the national registers of arthroplasty showed that about 12% of hip and knee arthroplasty undergo revision within 10 years after the primary surgery. The leading cause of hip revisions is aseptic loosening of components, knee joint periprosthetic infection (PPI). Some of the infectious complications, including those related to mechanical causes, remain out of sight. The aim of the study was to identify the frequency of unexpected infections during revision knee and hip arthroplasty performed for aseptic complications of any etiology. Materials and Methods. 839 cases of revision arthroplasty of knee and hip joints were analyzed, including 485 aseptic revisions in 450 patients. Clinical, X-ray, laboratory (complete blood count and comprehensive metabolic panel, coagulation panel) methods, synovial fluid analysis and microbiological examination of punctures, including intraoperative ones, were used. The ICM and EBJIS (European Bone and Joint Infections Society) consensus recommendations were used as criteria for assessing the presence of infection. Results. The average age of patients at the time of the revision was 61.7 years. The hip joint prevailed (59.4%), knee joint 40.6%. The growth of microorganisms in the intraoperative biomaterial was detected in 2.08% of observations: in 10 out of 287 patients after aseptic revision of the hip joints and in none of the 198 revisions of the knee joints. In 8 out of 10 cases, the causative agents were coagulase-negative staphylococci, including 6 MRSE; in two cases, anaerobic bacteria. All revisions were carried out by a one-stage method. Patients with detected PPI underwent systemic antibacterial therapy. At the stage of catamnesis, reinfection was assumed in one of the 10 identified cases of PPI, the patient did not show up for revision. In control 63% of the group of the other (aseptic) 470 patients, PPI developed in 4 cases, two-stage revisions were carried out. Conclusions. The frequency of infections accidentally detected during aseptic revisions of large joints was 2.08%. Three-time examination of joint punctures, including intraoperative, provides additional opportunities for the diagnosis of PPI during aseptic revision, and also allows you to choose the optimal stage of revision treatment. The experience gained makes it possible in certain cases to perform one-stage revision in the treatment of PPI.

“Unexpected” Infections in Revision Arthroplasty for Aseptic Loosening

Background. Data from the national registers of arthroplasty showed that about 12% of hip and knee arthroplasty undergo revision within 10 years after the primary surgery. The leading cause of hip revisions is aseptic loosening of components, knee joint periprosthetic infection (PPI). Some of the infectious complications, including those related to mechanical causes, remain out of sight. The aim of the study was to identify the frequency of unexpected infections during revision knee and hip arthroplasty performed for aseptic complications of any etiology. Materials and Methods. 839 cases of revision arthroplasty of knee and hip joints were analyzed, including 485 aseptic revisions in 450 patients. Clinical, X-ray, laboratory (complete blood count and comprehensive metabolic panel, coagulation panel) methods, synovial fluid analysis and microbiological examination of punctures, including intraoperative ones, were used. The ICM and EBJIS (European Bone and Joint Infections Society) consensus recommendations were used as criteria for assessing the presence of infection. Results. The average age of patients at the time of the revision was 61.7 years. The hip joint prevailed (59.4%), knee joint 40.6%. The growth of microorganisms in the intraoperative biomaterial was detected in 2.08% of observations: in 10 out of 287 patients after aseptic revision of the hip joints and in none of the 198 revisions of the knee joints. In 8 out of 10 cases, the causative agents were coagulase-negative staphylococci, including 6 MRSE; in two cases, anaerobic bacteria. All revisions were carried out by a one-stage method. Patients with detected PPI underwent systemic antibacterial therapy. At the stage of catamnesis, reinfection was assumed in one of the 10 identified cases of PPI, the patient did not show up for revision. In control 63% of the group of the other (aseptic) 470 patients, PPI developed in 4 cases, two-stage revisions were carried out. Conclusions. The frequency of infections accidentally detected during aseptic revisions of large joints was 2.08%. Three-time examination of joint punctures, including intraoperative, provides additional opportunities for the diagnosis of PPI during aseptic revision, and also allows you to choose the optimal stage of revision treatment. The experience gained makes it possible in certain cases to perform one-stage revision in the treatment of PPI.

A comparison of short term complication rate between 44 two- and 385 one-stage septic exchange arthroplasties in chronic periprosthetic joint infections

BACKGROUND

Periprosthetic joint infection is a serious disease with severe consequences for the patient's life. It is not known whether one-stage or two-stage replacement arthroplasty is superior. So far, there have been no studies on short-term complications between both surgical procedures. Therefore, we performed a retrospective study aiming to determine: 1) Does two-stage septic exchange arthroplasty have a higher short-term complication rate than one-stage? 2) Is there a difference in length of hospital stay and the in-hospital mortality rate between two- and one-stage septic exchange arthroplasty?

HYPOTHESIS

Two-stage septic exchange arthroplasty has a higher short-term complication rate than one-stage.

PATIENTS AND METHODS

Retrospectively 429 patients who underwent a one- or two stage revision arthroplasty (288 hips and 141 knees) due to chronic PJI between January 2015 and December 2016 were recruited (one-stage: n=385, two-stage: n=44). Differences in postoperative complications, surgical therapy, ASA, Charlson Comorbidity Index (CCI), length of hospital, need for plastic surgery, pathogen, in-hospital morbidity and in-hospital mortality were compared using multiple logistic and Poisson regression. Regarding comorbidities, age, gender and BMI the groups were comparable.

RESULTS

It was 3.5 times more likely to suffer from a medical complication if a two stage septic exchange was performed (OR 3.465, (95% CI: 2.573-4.358) [p<0.01]). In medical complications the two-stage group showed significantly more events (two-stage: 9 of 44=20.5%; one-stage: 30 of 385=7.8% [p=0.013]). The one-stage group showed more (not significant) nerve palsies after hip septic exchange. There was no difference in mortality between the groups (two-stage: 1/44=2.3%; one-stage: 3/385=0.8% [p=0.882]), the overall mortality was 0.93%. The duration of hospital stay was 23.9±19 days for the one-stage and 42.2±17.7 days for the two-stage group (p<0.001).

CONCLUSION

We found that it is more likely to develop a medical complication after two-stage septic revision arthroplasty. There was no difference in overall surgical complications between the two surgical approaches. We did not find a significant difference in short term mortality between the two techniques, although the mortality rate was slightly higher in the two-stage group.

LEVEL OF EVIDENCE

III, retrospective case control study.

Results at 10-24 years after single-stage revision arthroplasty of infected total hip arthroplasty in patients under 45 years of age

INTRODUCTION

The management of periprosthetic joint infection (PJI) of the hip is a challenging procedure. One-stage exchange is carried out in specialist centres with comparable infection free survival rates. However, there is a paucity of long-term results of this approach in young patients.

METHODS

All patients undergoing one-stage exchange due to PJI with a known causative organism ⩽45 years of age with a minimum of 10-year follow-up (mean 15 years; range 10-24 years) were enrolled in this retrospective study. Patients older than 45 years of age or patients with a two-stage procedure were excluded from the study. The primary outcome measure was failure rate with special focus on reinfection at latest follow-up. Failure was defined as revision surgery for infection. Furthermore, functional outcome using the Harris Hip Score was determined initially and at latest follow-up. 26 patients fulfilled the inclusion criteria and were available for final follow-up.

RESULTS

The study cohort consists of 16 male and 10 female patients with a mean age of 36.8 years (range 20-45 years) and a mean BMI of 29.8 kg/m2 (range 20.7-40.6 kg/m2). Prior to the single-stage procedure, the patients underwent an average of 3.1 previous surgical interventions (range 1-9). The mean hospital stay after septic revision was 19.7 days (11-33 days). Most infections were caused by Staphylococcus epidermidis (n = 8, 30.8%), followed by Staph. aureus (n = 7, 26.9%) and Propionibacterium acnes (n = 6, 23.1%). At latest follow-up, the overall survival rate was 76.9 %, while infection control could be achieved in 96.2%. At final examination, the mean Harris Hip Score improved from 46.2 to 78.9 (range 18.0-99; SD, 22.6).

CONCLUSIONS

Single-stage revision surgery for the management of PJI in patients ⩽45 years is a successful treatment option with high infection control, even after long-term follow-up.

Are Cement Spacers and Beads Loaded with the Correct Antibiotic(s) at the Site of Periprosthetic Hip and Knee Joint Infections?

The optimal impregnation of antibiotic-loaded bone cement in the treatment of periprosthetic hip and knee joint infection is unknown. It is also unclear, whether a suboptimal impregnation might be associated with a higher persistence of infection. A total of 93 patients (44 knee, 49 hip) were retrospectively evaluated, and the most common organism was a methicillin-resistant Staphylococcus epidermidis, followed by methicillin-susceptible Staphylococcus aureus. Of all the organisms, 37.1% were resistant against gentamicin and 54.2% against clindamycin. All organisms were susceptible against vancomycin. In 41 cases, gentamicin-loaded beads were inserted and in 52 cases, spacers: (2 loaded only with gentamicin, 18 with gentamicin + vancomycin, 19 with gentamicin + clindamycin, and 13 with gentamicin + vancomycin + clindamycin). The analysis of each antibiotic impregnation showed that complete susceptibility was present in 38.7% of the cases and partial susceptibility in 28%. In the remaining 33.3%, no precise statement can be made because either there was a culture-negative infection or the antibiotic(s) were not tested against the specific organism. At a mean follow-up of 27.9 months, treatment failure was observed in 6.7% of the cases. Independent of which antibiotic impregnation was used, when the organism was susceptible against the locally inserted antibiotics or not tested, reinfection or persistence of infection was observed in the great majority of cases. Future studies about the investigation of the optimal impregnation of antibiotic-loaded bone cement are welcome.

Positive Microbiological Findings at the Site of Presumed Aseptic Revision Arthroplasty Surgery of the Hip and Knee Joint: Is a Surgical Revision Always Necessary?

Little is known about patients that undergo presumed aseptic revision arthroplasty surgery of the hip and knee joint and having positive microbiological findings of the intraoperatively taken tissue samples. 228 “aseptic” operations were retrospectively analyzed from prospectively collected data with regard to the following parameters: demographic data; reasons for primary and revision surgery, respectively; time between primary and revision surgery; preoperative laboratory findings; microbiological and histopathological findings; type and length of systemic antibiotic therapy; clinical outcome; and follow-up. Identification of microorganisms was present in 8.8% of the cases (9.3% of the hip and 7.8% of the knee cases). Preoperatively, the median CRP value was 8.4 mg/l (normal values 0-5.0 mg/l) and the median WBC count 8,100 × 10⁶/l (normal values 3, 700‐10,100 × 10⁶/l). The most common identified organism was methicillin-resistant Staphylococcus epidermidis in 30%, followed by viridans streptococci in 15% of the cases. In 7 cases, the microbiological findings were interpreted as a contamination, and no antibiotic therapy was administered. In the other cases, a systemic antibiotic therapy was applied for a time period between 2 weeks and 3 months. 68.4% of the patients did not have any infectious complications at a median follow-up of 20 (3-42) months. The present study indicates that more than 2/3 of the cases with positive microbiological findings at the site of presumed aseptic revision arthroplasty surgery of the hip and knee joint can be successfully treated conservatively and they do not require any further surgical therapy.

The Impact of α-Defensin Test in Diagnosing Periprosthetic Infection After Total Ankle Arthroplasty

Periprosthetic joint infection (PJI) after total ankle arthroplasty (TAA) is a serious complication, and a reliable diagnostic test to identify PJI is needed. The purpose of this study was to investigate the use of synovial α-defensin levels in identifying PJI of the ankle. Data from 33 patients were retrospectively collected between September 2015 and May 2018. Patients who had pain or suspected loosening after TAA and who had undergone joint aspiration were included in the study. Aspiration was performed in a semisterile theatre. Synovial fluid was processed in descending order for microbiological cultures, α-defensin, leukocyte esterase strip test, and cell count. A periprosthetic infection was defined by Musculoskeletal Infection Society criteria. The sensitivity, specificity, and overall accuracy were calculated, and based on a receiver operating characteristic curve, the quality of the α-defensin test was determined. The calculated area under the curve was 0.97 ± 0.32. Two of 33 patients fulfilled the 2014 Musculoskeletal Infection Society criteria and were scheduled for septic revision arthroplasty. Sensitivity, specificity, and overall accuracy of the α-defensin test were 100% (95% confidence interval [CI], 15.8% to 100%), 93.5% (95% CI, 78.6% to 99.2%), and 93.9% (95% CI, 79.8% to 99.3%), respectively. The positive predictive value was 50% (95% CI, 20.7% to 79.3%), and the negative predictive value was 100%. The α-defensin test seems to be the best available synovial test to detect a late-onset PJI after total ankle arthroplasty. Further prospective studies with a larger number of patients are required.

What are the Factors Associated with Re-revision After One-stage Revision for Periprosthetic Joint Infection of the Hip? A Case-control Study

BACKGROUND

Despite increased interest in one-stage revision arthroplasty for periprosthetic joint infection (PJI) of the hip, the reported infection-free proportions after the one-stage approach are still comparable to that of two-stage revision. However, we still lack studies that analyze factors associated with any re-revision after one-stage revision arthroplasty for PJI.

QUESTIONS/PURPOSES

After one-stage revision arthroplasty for PJI of the hip, what factors were associated with an increased risk of re-revision, and what factors were associated with an increased risk of reinfection?

METHODS

We performed a single-center retrospective case-control analysis. Patients who underwent one-stage revision arthroplasty for PJI of the hip in our hospital between 2009 and 2017, and who were subsequently re-revised due to any reason, including reinfection, were identified from our electronic medical records. A total of 121 patients who underwent repeat revision for any reason after one-stage exchange for PJI of the hip were identified. The re-revision procedures were performed after a mean of 407 days. The primary reasons for re-revision were repeated hip dislocation in 53 of 121 patients (44%), reinfection including both new and persistent infections in 40 of 121 patients (33%), and aseptic loosening in 16 patients (13%). Forty-three patients underwent another revision procedure after the re-revision procedure (43 of 121; 36%). More than 40 potential patient-, joint- and surgery-related risk factors were investigated and compared with a 1:1 matched control participants by age, sex and year of the one-stage revision. Similar to the re-revision patients, controls were treated for PJI with one-stage revision arthroplasty; however, they did not undergo subsequent revision for any reason by the latest followup examination. The mean followup of the control group was 66 months (range, 17-119 months). The mean length of hospital stay was 26 days in the re-revised group (SD, 11.6 days) compared with 22 days for the controls (SD, 6.6 days). All analyses were performed to identify factors associated with general re-revision and reinfection.

RESULTS

The independent factors associated with repeat revision for any reason were persistent wound drainage for at least 1 week (odds ratio [OR], 7.4; 95% CI, 2.6-20.6; p < 0.001), isolation of enterococci (OR, 4.8; 95% CI, 1.4-15.7; p = 0.010), and prior surgery due to infection before the one-stage hip revision (OR, 3.6; 95% CI, 2.0-6.4; p < 0.001). The factors associated with reinfection including both new and persistent infections were prolonged wound drainage (OR, 6.9; 95% CI, 2.2-21.5; p = 0.001) and prior surgery due to infection (OR, 4.3; 95% CI, 1.9-9.5; p < 0.001).

CONCLUSIONS

Prolonged wound drainage after the one-stage revision arthroplasty for PJI of the hip must be treated rigorously. Patients with a history of a prior surgical procedure due to hip infection should be informed about the risk of further re-revision when deciding for the one-stage exchange. In case of enterococcal isolation, surgeons may consider another treatment approach rather than the one-stage exchange. Furthermore, we recommend the use of dual mobility cups when performing the one-stage revision hip arthroplasty to reduce the risk of dislocation.Level of Evidence Level III, therapeutic study.

Partial two-stage exchange at the site of periprosthetic hip joint infections

INTRODUCTION

In the past 10 years an increasing number of studies about partial two-stage exchange arthroplasty in the management of periprosthetic hip infections have been published. The aim of the present work was to systematically review the current knowledge about this procedure, and critically verify the success as well as the complications of this treatment option.

MATERIALS-METHODS

A literature search was performed through PubMed until June 2018. Search terms were "partial two stage hip" and "partial retention hip", and "retaining well fixed hip".

RESULTS

A total of 7 studies reporting on a total of 80 patients could be identified. All studies had a level of evidence IV. The great majority of the studies reported on the isolated removal of the acetabular cup and placement of an antibiotic-loaded cement spacer head onto the retained, well-fixed stem. Most of the periprosthetic infections were caused by staphylococci. The infection eradication rate varied between 81.3 and 100% at a mean follow-up between 19 and 70 months. Poor outcome was observed at the site of MRSA infections.

CONCLUSIONS

The partial two-stage exchange arthroplasty appears to be a possible option in the management of PJI when one prosthetic component is well-fixed so that their removal might result in significant bone loss and compromise of fixation at the time of the later prosthesis reimplantation, and the causative organisms are not multiresistant. The small numbers published about this protocol does not allow for a generalization of application and should be only applied in highly selected patients. Future studies with larger collectives and longer follow-ups are welcome to evaluate the clinical success of this option and its possible role in the management of PJI.

Resection Arthroplasty Treatment Possibility for a Patient with Fibrous Dysplasia, Chronic Osteomyelitis, and Recurrent Periprosthetic Hip Joint Infection: A Case Report

Introduction:

Fibrous dysplasia (FD) is a rare congenital abnormality, in which parts of bone are replaced with fibrous stroma and woven bone, making it susceptible to pathological fractures. Osteomyelitis following intramedullary nailing of such fractures of the femur and periprosthetic joint infection (PJI) after hip arthroplasty is one of the most devastating complications and a true challenge for the orthopedic surgeon. To the best of our knowledge, this is the 1st time, such complications are reported in a patient with monostotic FD.

Case Report:

We present a 30-year-old male patient with FD of the left femur, chronic osteomyelitis, recurrent periprosthetic hip joint infection after multiple revision arthroplasties, and two episodes of axillary vein thrombosis. Due to the severe structural changes and a large medial wall defect in the proximal femur with impairment of a large soft tissue zone, it was decided during operation not to put any other implants in the hip joint and perform only a resection hip arthroplasty procedure.

Conclusion:

Managing a recurrent PJI in a patient with poor bone quality, severe bone defects, soft tissue compromise, and thromboembolic events is challenging even for the most experienced orthopedic surgeon. Resection arthroplasty is a salvage treatment option that should be considered in such complex cases.

Management of Periprosthetic Joint Infection

Periprosthetic joint infection (PJI) is a serious complication after arthroplasty, which is associated with pain, prolonged hospital stay, multiple surgeries, functional incapacitation, and even mortality. Using scientific and efficient management protocol including modern diagnosis and treatment of PJI and eradication of infection is possible in a high percentage of affected patients. In this article, we review the current knowledge in epidemiology, classification, pathogenesis, diagnosis and treatment of PJI.

Tertiary care centre adherence to unified guidelines for management of periprosthetic joint infections: a gap analysis

BACKGROUND

The success rate of surgical treatment for periprosthetic joint infection (PJI) remains inconsistent in the literature. Variability in PJI clinical guidelines and surgeon adherence to guidelines could affect treatment success. The objectives of this study were to appraise current recommendations for PJI management and develop a unified clinical standard of care, to perform a gap analysis of PJI cases in a tertiary institution to determine the rate of guideline adherence, and to determine if adherence to unified PJI guidelines affected 2-year treatment outcomes.

METHODS

We appraised the PJI guidelines from 3 academic medical societies, and consistent statements were aggregated. We retrospectively reviewed all PJI cases in a tertiary care institution. We defined PJI based on Musculoskeletal Infection Society PJI criteria. Surgeon adherence to preoperative, intraoperative, surgical and medical management guidelines was calculated, and we evaluated the association between guideline adherence and 2-year treatment outcomes.

RESULTS

The institutional rate of PJI was 1.13% (38 of 3368). Treatment success was 57.8% at 2 years. Unified guideline adherence percentages varied substantially: 92% of patients had preoperative erythrocyte sedimentation rate and C-reactive protein, 97% had intraoperative tissue cultures, 42% had appropriate preoperative arthrocentesis, and 74% underwent guideline-appropriate surgery. Performing appropriate preoperative arthrocentesis significantly correlated with positive treatment outcomes at 2 years (p = 0.028).

CONCLUSION

Adherence to PJI guidelines varies considerably, indicating that clinicians are either unaware of them or do not recognize their value for PJI treatment. This study shows the need for institution-based PJI treatment pathways that are consistent with published guidelines and the need to monitor adherence.

Nano-carrier based drug delivery systems for sustained antimicrobial agent release from orthopaedic cementous material

Total joint replacement (TJR), such as hip and knee replacement, is a popular procedure worldwide. Prosthetic joint infections (PJI) after this procedure have been widely reported, where treatment of such infections is complex with high cost and prolonged hospital stay. In cemented arthroplasties, the use of antibiotic loaded bone cement (ALBC) is a standard practice for the prophylaxis and treatment of PJI. Recently, the development of bacterial resistance by pathogenic microorganisms against most commonly used antibiotics increased the interest in alternative approaches for antimicrobial delivery systems such as nanotechnology. This review summarizes the efforts made to improve the antimicrobial properties of PMMA bone cements using nanotechnology based antibiotic and non-antibiotic delivery systems to overcome drawbacks of ALBC in the prophylaxis and treatment of PJIs after hip and knee replacement.

One-Stage Revision for Infected Total Hip Arthroplasty

Prosthetic joint infection (PJI) is a challenging complication following total hip arthroplasty (THA). Two-stage exchange arthroplasty is preferred for treating chronic PJI of THA, although specialized centers have reported comparable outcomes with protocol-based, 1-stage exchange arthroplasty. A main requirement is presurgical determination of the infecting organism's sensitivity. The therapeutic goal is control of the infection and maintenance of joint function. It offers advantages, including a single operative procedure, fewer antibiotics, and reduced hospitalization time and relative overall costs.