Topical Vancomycin Powder and Dilute Povidone-Iodine Lavage Reduce the Rate of Early Periprosthetic Joint Infection After Primary Total Knee Arthroplasty

BackgroundVancomycin powder and dilute povidone-iodine lavage (VIP) was introduced to reduce the incidence of periprosthetic joint infection (PJI) in high-risk total knee arthroplasty (TKA) patients. We hypothesize that VIP can reduce the incidence of early PJI in all primary TKA patients, regardless of preoperative risk.MethodsAn infection database of primary TKAs performed before a VIP protocol was implemented (January 2012-December 2013), during a time when only high-risk TKAs received VIP (January 2014-December 2015), and when all TKAs received VIP (January 2016-September 2019) at an urban, university-affiliated, not-for-profit orthopedic hospital was retrospectively reviewed to identify patients with PJI. Criteria used for diagnosis of PJI were the National Healthcare Safety Network and Musculoskeletal Infection Society guidelines.ResultsVIP reduced early primary TKA PJI incidence in both the high-risk and all-risk cohorts compared with the pre-VIP cohort by 44.6% and 56.4%, respectively (1.01% vs 0.56% vs 0.44%, P = .0088). In addition, after introducing VIP to all-risk TKA patients, compared with high-risk TKA patients, the relative risk of PJI dropped an additional 21.4%, but this finding did not reach statistical significance (0.56% vs 0.44%, P = .4212). There were no demographic differences between the 3 VIP PJI cohorts.ConclusionVIP is associated with a reduced early PJI incidence after primary TKA, regardless of preoperative risk. With the literature supporting its safety and cost-effectiveness, VIP is a value-based intervention, but given the nature of this historical cohort study, a multicenter randomized controlled trial is underway to definitively confirm its efficacy.     

Risk of Gastrointestinal Bleeding With Extended Use of Nonsteroidal Anti-Inflammatory Drug Analgesia After Joint Arthroplasty

BackgroundChronic nonsteroidal anti-inflammatory drug (NSAID) use is associated with gastrointestinal bleeding via inhibition of endogenous mucosal protection and platelet aggregation. This study aimed to determine whether extended NSAIDs after joint arthroplasty is associated with increased risk of gastrointestinal bleeding.MethodsThis was a retrospective study examining 28,794 adults who underwent joint arthroplasty by one of 50 surgeons from 2016 to 2018. Episodes of gastrointestinal bleeding within 90 days postoperatively were identified prospectively. Postoperative medications were reported directly by patients with electronic questionnaires. The primary analysis was performed using binary logistic regression.ResultsA total of 74 (0.26%) episodes of gastrointestinal bleeding occurred within 90 days (median 8 days) postoperatively. Of 5086 patients with complete data included in the primary analysis, 59.6% had used NSAIDs with median duration of 2 weeks (interquartile range, 0-6 weeks). Patients with gastrointestinal bleeding were significantly older (71.3 vs 67.0 years), required longer hospitalizations (2.1 vs 1.5 days), and more commonly had a history of peptic ulcers (10.8% vs 0.9%). However, there was no positive association between NSAID use and gastrointestinal bleeding. In fact, the odds of gastrointestinal bleeding were lower in patients taking NSAIDs. Gastrointestinal bleeding was associated with anticoagulants, antiplatelet agents, and, to a lesser extent, aspirin.ConclusionNSAIDs were not associated with gastrointestinal bleeding and may be prescribed safely for a majority of patients after joint arthroplasty. The greatest odds of gastrointestinal bleeding occurred in patients with peptic ulcer disease and those who received antiplatelet and anticoagulation agents. Increasing age and bilateral surgery were also associated with gastrointestinal bleeding.Level of EvidenceLevel III.     

Assessment of Informal Caregiver Assistance and Strain With Total Hip and Knee Arthroplasty

BackgroundThe purpose of our study is to assess which patient-related and caregiver-related factors are predictive of caregiver strain and assistance when caring for total hip and knee arthroplasty (THA and TKA) patients within 2 weeks after surgery.MethodsWe conducted a prospective study of caregivers of participants enrolled in 2 randomized trials. Caregivers provided demographics and completed the Caregiver Strain Index and Caregiver Assistance Scale pre-surgery and post-surgery. We performed backwards stepwise regression with mixed-effects negative binomial models to investigate predictors of caregiver strain and assistance for THA and TKA caregivers.ResultsThree hundred six caregiver/patient pairs were included. Our models of caregiver strain found Caregiver Assistance Scale scores and patient age to be predictive for all caregivers. We also found caregiver gender and smoking status to be predictive for THA caregivers and caregiver age to be predictive for TKA caregivers. Our models of assistance provided by caregivers found time (post-surgery vs pre-surgery) was predictive for all caregivers. We also found patient body mass index, and patient and caregiver gender to be predictive for THA caregivers, and patient and caregiver employment status and caregiver education level to be predictive for TKA caregivers.ConclusionOur study identifies patient-related and caregiver-related factors which are associated with caregiver strain and assistance when caring for arthroplasty patients. As this is the first study to assess assistance provided by caregivers, it is important for future research to validate our results and to further explore whether patient-reported outcomes may also be related to assistance and strain.     

Intraosseous Regional Administration of Vancomycin in Primary Total Knee Arthroplasty Does Not Increase the Risk of Vancomycin-Associated Complications

BackgroundPeriprosthetic joint infection (PJI) after total knee arthroplasty (TKA) is a rare but major complication. Owing to an increasing antibiotic resistance in bacteria causing PJI, vancomycin has been investigated as a prophylactic agent. Intraosseous regional administration (IORA) of vancomycin achieves significantly higher local tissue concentrations than systemic administration. There are limited data on IORA of vancomycin with respect to vancomycin-associated complications.MethodsSingle-surgeon retrospective review of primary TKA was performed between January 2015 and May 2019. All patients received 500 mg of IORA of vancomycin after tourniquet inflation and 3 × 1 g intravenous cefazolin in 24 hrs. Preoperative data collected included age, gender, body mass index, American Society of Anesthesiologists (ASA) score, diabetes, and chronic kidney disease (CKD). We documented in-hospital complications and complications requiring readmission within 12 months. Primary outcome measures were the incidence of acute kidney injury (AKI), ‘red man syndrome’ (RMS), and neutropenia. The secondary outcome measure was PJI incidence.ResultsWe identified 631 primary TKAs in 556 patients, of which 331 received IORA. The mean age was 67.7 ± 8.7 years, and 57.8% were women. CKD was prevalent in 17.2% of the cohort. AKI occurred in 25 (3.9%) cases. After controlling for covariates, CKD was the only significant predictor of AKI (odds ratio = 3.035, P = .023). RMS and neutropenia were not observed in this cohort. The 90-day PJI rate was 0%, and the 1-year PJI rate was 0.2%.ConclusionsLow-dose IORA of vancomycin in addition to standard intravenous systemic cefazolin prophylaxis in TKA is safe without significant adverse effects of vancomycin such as AKI, RMS, or neutropenia.     

A Web-Based Interactive Patient-Provider Software Platform Does Not Increase Patient Satisfaction or Decrease Hospital Resource Utilization in Total Knee and Hip Arthroplasty Patients in a Single Large Hospital System

BackgroundWeb-based platforms used to enhance patient-provider communication are being explored to improve patient satisfaction and care delivery, and decrease cost. This study tested a web-based interactive patient-provider software platform (IPSP), JointCOACH, which enabled patient communication with their care team and preparatory/recovery guidance. The aims of this study are to compare (1) patient satisfaction and (2) healthcare resource utilization by patients who underwent total knee and hip replacements and added IPSP to standard of care (SOC).MethodsThis study is a prospective, randomized clinical trial at a single large academic healthcare system. Between May 2018 and March 2020, 399 patients undergoing elective total hip or knee arthroplasty were randomized to SOC arm (n = 204) or SOC + IPSP arm (n = 195). Patient demographics, surgical details, and comorbidities were collected. Patient satisfaction was assessed using Visual Analog Scale and the Picker Patient Experience-15. Healthcare utilization was measured using length of stay, emergency department and office visits, office calls, readmissions, and reoperations at 30 and 90 days after surgery.ResultsNo difference was found in length of stay between SOC and SOC + IPSP. No differences were found in 30-day or 90-day satisfaction or in healthcare resource utilization (P > .05) including number of office and emergency department visits, phone calls, and readmissions.ConclusionStatistical differences were not found in satisfaction and healthcare utilization with the addition of IPSP to SOC. IPSP can be used to reinforce patient education and communication between the patient and provider, and should be evaluated as an element of virtual care rather than supplementing traditional in-office follow-up.Clinicaltrials.govMore information on this study can be found at clinicaltrials.gov NCT03499028.     

Early Identification of Patient Satisfaction Two Years After Total Knee Arthroplasty

BackgroundThere are numerous reports of poor satisfaction after total knee arthroplasty (TKA), yet there is little known about when to use evidence-based models of care to improve patient outcomes.ObjectiveThis study aimed to characterize longitudinal changes in patient-reported satisfaction after TKA and to identify factors for early identification of poor satisfaction.MethodsFor a cohort of primary TKA surgeries (n = 86), patient-reported outcomes were captured one week before TKA and 6 weeks, 12 weeks, 6 months, and 1 and 2 years after TKA. “Satisfied” versus “not fully satisfied” patients were defined using a binary response (≥90 vs <90) from a 100-point scale. Wilcoxon signed-rank tests identified changes in satisfaction between follow-up times, and longitudinal analyses examined demographic and questionnaire factors associated with satisfaction.ResultsImprovements in satisfaction occurred within the first 6 months after TKA (P ≤ 0.01). Preoperative patient-reported outcome measures alone were not predictive of satisfaction. Key factors that improved longitudinal satisfaction included higher Oxford Knee Scores (odds ratio (OR) = 2.1, P < .001), general health (EQ-VAS, OR = 1.3, P = .03), and less visual analog scale pain (VAS; OR = 1.7, P < .001). Differences in these factors between satisfied and not fully satisfied patients were identified as early as 6 weeks after surgery.ConclusionVisibly different satisfaction profiles were captured among satisfied and not fully satisfied patient responses, with differences in patient-perceived joint function, general health, and pain severity occurring as early as 6 weeks after surgery. This study provides metrics to support early identification of patients at risk of poor TKA satisfaction, enabling clinicians to apply timely targeted treatment and support interventions, with the aim of improving patient outcomes.     

Is Patient Satisfaction After Total Knee Arthroplasty Predictable Using Patient Characteristics and Preoperative Patient-Reported Outcomes?

BackgroundDissatisfaction after total knee arthroplasty (TKA) remains a difficult problem. Patient characteristics and preoperative patient-reported outcomes (PROs) are potential predictors of satisfaction one year after TKA. Being able to predict the outcome preoperatively might reduce the number of less satisfied patients.MethodsA retrospective cohort study on prospectively collected data of 1239 primary TKA patients (ASA I-II, BMI <35) was performed. Primary outcome was degree of patient satisfaction one year after TKA (Numeric Rating Scale (NRS) 0-10). Secondary outcomes were degree of patient satisfaction six months and two years after TKA and being dissatisfied (NRS 0-6) or satisfied (NRS 7-10) at all three time points. Multivariate linear and binary logistic regression analyses were executed with patient characteristics and preoperative PROs as potential predictors.ResultsOne year after TKA, median NRS satisfaction score was 9.0 (8.0-10.0) and 1117 (90.2%) patients were satisfied. BMI, degree of medial cartilage damage, previous knee surgery, Knee injury and Osteoarthritis Outcome Score-Physical Function Short Form score, EQ VAS score, and anxiety were identified as predictors of the degree of patient satisfaction (P = .000, R² = 0.027). Models on secondary outcomes reported R² of 1.7%-7.1% (P < .05). All models showed bad agreement between observed and predicted values for lower NRS satisfaction scores and being dissatisfied.ConclusionThe degree of patient satisfaction and the chance of being dissatisfied or satisfied six months, one, and two years after TKA are predictable by patient characteristics and preoperative PROs but not at a reliability level that is clinically useful.     

Septic Revision Total Knee Arthroplasty Is Associated With Significantly Higher Mortality Than Aseptic Revisions: Long-Term Single-Center Study (1254 Patients)

BackgroundThe aim of this study is to examine the differences in long-term mortality rates between septic and aseptic revision total knee arthroplasty (rTKA) in a single specialist center over 17-year period.MethodsRetrospective consecutive study of all patients who underwent rTKA at our tertiary center between 2003 and 2019 was carried out. Revisions were classified as septic or aseptic. We identified patients’ age, gender, American Society of Anesthesiologists grade, and body mass index. The primary outcome measure was all-cause mortality at 5 years, 10 years, and over the whole study period of 17 years. Death was identified through both local hospital electronic databases and linked data from the National Joint Registry/NHS Personal Demographic Service. Kaplan-Meier survival curves were used to estimate time to death.ResultsIn total, 1298 consecutive knee revisions were performed on 1254 patients (44 bilateral revisions) with 985 aseptic revisions in 945 patients (75.4%) and 313 septic revisions in 309 patients (24.6%). Average age was 70.6 years (range 27-95) with 720 females (57.4%). Septic revisions had higher mortality rates; patients’ survivorship for septic vs aseptic revisions was 77.6% vs 89.5% at 5 years, 68.7% vs 80.2% at 10 years, and 66.1% vs 75.0% at 17 years; these differences were all statistically significant (P < .0001). The unadjusted 10-year risk ratio of death after septic revision was 1.59 (95% confidence interval 1.29-1.96) compared to aseptic revisions.ConclusionrTKA performed for infection is associated with significantly higher long-term mortality at all time points compared with aseptic revision surgery.Level of EvidenceLevel IV.     

Can Met Expectations Moderate the Relationship Between Pain/Function and Satisfaction in Total Knee Arthroplasty?

BackgroundApproximately 20% of patients are dissatisfied with their total knee arthroplasty (TKA) at 1-year post-surgery. Met expectations have been found by some to significantly predict satisfaction. The role of met expectations in determining patient satisfaction has not been exhaustively explored. The primary aim of this study is to evaluate if met expectations moderate the relationship between pain and function variables and satisfaction.MethodsPatients who underwent primary TKA for osteoarthritis were included in the study (n = 304). Patient-reported outcomes at pre-surgery and 1-year post-surgery were collected. The Knee Society Score (KSS) satisfaction subscale was used as the dependent variable. Candidate independent variables included the following: demographics, KSS, Knee injury and Osteoarthritis Outcome Score (KOOS), 12-Item Short Form Health Survey (SF-12), Hospital Anxiety and Depression Scale, Pain Catastrophizing Scale, Self-Administered Comorbidity Questionnaire, and University of California Los Angeles activity score. Separate linear regression models were created to test interactions for KSS met expectations with pain and KSS met expectations with function.ResultsSignificant predictors of satisfaction were KSS symptoms (pain), KOOS activities of daily living (function), KSS met expectations, KOOS pre-surgery activities of daily living, body mass index, and SF-12 general health. A significant interaction between met expectations and pain was found (P = .043) and the met expectations and function interaction approached significance (P = .086). For both interactions, as met expectations increased, pain and function predicted satisfaction less strongly.ConclusionMet expectations were found to moderate the relationship between pain and satisfaction. There may be more value in improving pain for patients with low met expectations.Level of evidenceLevel IV.