More than one million Americans undergo knee, hip, elbow and shoulder replacements annually. About one in 75 patients, less than two percent, develop bacterial infections around the joint implants, called prosthetic joint infections. Though the numbers of affected patients are small, the consequences of prosthetic joint infection can be severe.

HSS performs more joint replacement surgeries than any other hospital in the world—a total of 11,000 annually. “Our infection rate is significantly below the national average, but prosthetic joint infections are devastating for affected patients,” says Mathias P. Bostrom, MD, chief of the Adult Reconstruction and Joint Replacement Service at HSS. “Many do not return to the full potential of their original joint replacement.”

It is essential to identify the bacteria causing an infection so that a patient receives the right antibiotic. Since current blood tests only report levels of inflammatory markers, such as C-reactive protein and erythrocyte sedimentation rate, surgeons typically collect a fluid sample from the joint via needle aspiration. However, standard lab test results take at least three days and fail to identify the infectious pathogen in 15 to 20 percent of cases due to the challenges of growing bacteria in culture.

When bacterial identification is unknown, surgeons withhold antibiotics until they obtain tissue samples during the implant removal surgery. The removal procedure involves taking out the contaminated implant hardware and infected soft tissue. Surgeons insert a temporary spacer containing high-dose antibiotics, and they may also place antibiotic beads into the joint. Patients are also treated with high-dose intravenous antibiotics. It takes about six weeks to three months for an infection to clear before patients can undergo another operation to remove the spacer and receive a new permanent joint implant.

The idea for finding a new way to improve the diagnostic approach for patients with prosthetic joint infections began with principal investigator Laura Donlin, PhD, co-director of the Derfner Foundation Precision Medicine Laboratory and a member of the arthritis and tissue degeneration program of the HSS Research Institute. She attended a talk by a professor of bioengineering at Stanford University who had analyzed DNA shed from transplanted organs circulating in blood, called cell-free DNA, as an early detection system for transplant rejection. But through genomic sequencing of circulating cell-free DNA, he had also found increased levels of viral cell-free DNA in some patients’ blood indicating they had developed viral infections while taking immunosuppressant medications.

Dr. Donlin wondered if it would be possible to use the same approach to identify bacteria in patients with prosthetic joint infections. “It was unknown whether bacterial DNA from localized tissue infections around joint implants would be detectable in blood,” says Dr. Donlin. “It’s a challenging environment with many other bits of circulating microbial DNA from skin flora, the gut microbiome and other infections.”

For their proof-of-concept study, Dr. Donlin together with orthopedic surgeons including Michael P. Cross, MD and Dr. Bostrom and colleagues at HSS and Weill Cornell Medicine collected blood samples from 53 patients with known hip or knee prosthetic joint infections beginning in 2018. Karius, a genomic insights company based in Redwood City, California, sequenced blood samples collected from patients before treatment for infection and at the time of reimplantation surgery. They compared microbial cell-free DNA in the blood samples to their proprietary database of more than 1,300 known microbial genomes. The HSS investigators compared sequencing findings with results from standard tissue cultures.

Among surgical tissue samples from these 53 patients, traditional lab cultures identified the bacterial species in 35 cases and the bacterial genus in 11 cases, an overall detection rate of 87 percent. Microbial cell-free DNA sequencing identified the bacterial species in 23 cases in agreement with standard culture results. The new approach also identified the bacterial species in eight cases where cultures identified the bacterial genus only and four cases where cultures failed to determine the presence of bacteria.

On its own, microbial cell-free DNA sequencing pinpointed the bacterial species in 66 percent of samples. However, as an addition to standard culture results, it increased pathogen detection from 87 to 94 percent of samples. Microbial cell-free DNA sequencing was three days faster reporting the bacterial species for cases where culture results had only identified the bacterial family.

Analyses of follow-up blood samples after joint removal surgery and treatment with antibiotics showed undetectable or reduced bacteria levels. “For most patients, we saw microbial cell-free DNA levels drop below detection, indicating the infections had most likely cleared,” says Dr. Donlin. “But there were some cases with lower yet detectable levels after six weeks of antibiotic treatment. Considering that cell-free DNA lives in the bloodstream for only a few minutes, that meant these patients still had an ongoing infection and might require modification to their antibiotic treatment plan.”

“An indication that there is an infection, at least to the genus level for samples that fail to show results in standard cultures, will be beneficial for diagnosing and treating patients earlier for improved outcomes,” says Dr. Bostrom. “As a monitoring tool, microbial cell-free DNA sequencing has the potential to provide important information on the right time to change or stop antibiotics and for reimplantation surgery.”

Dr. Donlin and colleagues are improving the sensitivity and specificity of the new diagnostic method in collaboration with Karius and planning a multicenter study to test it on a larger scale. “In the future, we hope microbial cell-free DNA sequencing will prove to be a useful tool for detecting joint infections more quickly than is currently possible,” Dr. Donlin says. “We’re very excited that our innovation may one day translate to improved outcomes for patients.”

The study was supported by funding provided by the Price Family Foundation, the Stavros Niarchos Foundation Complex Joint Reconstruction Center at HSS, the Feldstein Medical Foundation, the Carson Family Charitable Trust, the Ambrose Monell Foundation, the HSS Research Institute’s David Z. Rosensweig Center for Genomics Research, the National Institutes of Health, the Bill and Melinda Gates Foundation, the Leukemia & Lymphoma Society and the National Science Foundation.

"In our study, we have shown no significant difference in terms of infection cure rate at a mean of two years postoperative, in our select group of patients, between the one-stage and two-stage revision," said lead study author Michael B. Cross, MD, an orthopedic surgeon specializing in hip and knee replacement at HSS. "This study provides good preliminary data to support a randomized trial and that is ongoing [in the U.S.]."

PJI is one of the most devastating and frequent complications after lower extremity joint replacement, often requiring revision surgery. "Infection is the hardest complication of joint replacement for both surgeon and patient," said Dr. Cross. For the patient, infection requires additional surgery and time off work, and there is a constant fear that the infection won’t be cured. Revision surgery can be tricky for surgeons.

Traditionally, surgeons have treated infected knee replacements with a two-stage approach, by removing the implants and inserting an antibiotic spacer to treat the local infection. After the knee is retested and free of infection, the knee implant is reinserted. The infection cure rate with this technique is 85%, and the process takes roughly three months at a minimum, which is a long time for patients to be out of work and not active. Another option used more frequently in Europe and for select cases in the United States, is a one-stage process, where the implant is removed, the knee joint is depleted of infected tissue and the knee replacement is put back in with a whole new set of sterile instruments. Critics of the one-stage technique have suggested that infection eradication is inferior to the historically successful two-stage revision. 

To shed light on the issue, researchers at HSS conducted a retrospective, case- control study to compare infection-free and all-cause revision survival between a matched cohort of one-stage and two-stage revisions for PJI. They identified 41 patients who underwent a one-stage revision for a PJI after a total hip or knee replacement through an IRB-approved database at HSS. Patients were matched (1:2) to controls treated with a two-stage revision, based on joint, sex, and body mass index.

With an average follow-up of 2.6 years, infection-free survival was not statistically or clinically significant, 80.5% in the one-stage group and 87.8% in the two-stage group (P=0.28). All-cause revision survival was 70.7% for one-stage and 75.6% for two-stage revisions (P=0.56), but 31.7% of two-stage patients had a history of revision in the affected joint compared to none of the one-stage patients (P<0.01). To control for this, survival was assessed for 41 one-stage and 56 two-stage patients with no history of revision in the affected joint. Infection-free survival (80.5% vs 92.9%; P=0.07) and all-cause revision survival (70.7% vs. 80.4%; P=0.27) were not different between subgroups. The reduced duration treatment and a single operation with the one-stage revision reduces cost and morbidity, in addition to being more convenient.

To the authors’ knowledge, the study is the first to compare one-stage and two-stage procedures, while controlling for previous replacement procedures and also the first to control patient population between two groups at a single institution. "It’s usually a hospital in Europe publishing data on a one-stage and a hospital in the United States publishing data on a two-stage, but obviously there is different bacteria and different hospital settings, so the comparison between the two is not as good," said Dr. Cross. "On average, there was a better success rate to treating the infections with the two-stage revision, but statistically it was not different between the two groups."

Dr. Cross noted that joint replacements are becoming more common, as the baby boomer generation ages and starts requiring more care for hips and knees. "The baby boomers are more active in their generation, and as a result of that, there is a huge influx in demand for hip and knee replacements," said Dr. Cross. "Knee replacements are growing exponentially, and they will continue that way until 2030."

In 2017, more than 47,000 Americans died after overdosing on opioids, such as heroin or fentanyl, a potent synthetic drug, government data show. Some 1.7 million more had a substance abuse disorder involving a prescription opioid. 

Patients who undergo joint replacement surgeries of all kinds conventionally have received an opioid prescription for their recovery—presenting critical opportunity for reducing the reliance on these medications and preventing abuse of the drugs. 

HSS is taking a multi-pronged approach to the problem, explained Michael P. Ast, MD, an orthopedic surgeon at the institution. HSS recently launched a hospital-wide initiative to reduce opioid use among its patients and to better understand when the powerful drugs are appropriate to prescribe, and in what quantities. The initiative has involved surgeons, anesthesiologists, nurses and other members of the care team.

A primary focus of the pain-management pathway is the emphasis on multimodal analgesia. The practice involves the combination of local anesthetics, nonsteroidal anti-inflammatory drugs, IV acetaminophen and, possibly but not necessarily, opioids. Multimodal analgesia for knee and hip replacement surgeries has been associated with a decrease in the use of opioids, according to the American Society of Anesthesiologists (https://www.asahq.org/about-asa/newsroom/news-releases/2018/03/multimodal-approach-to-pain-management-reduces-opioid-use).

Another aspect of the initiative involves research, and HSS currently has more than a dozen ongoing studies to assess alternatives to opioid analgesia. "We’re doing things like intraoperative and post-op acupuncture therapy and we’re looking at alternative treatments in the form of nerve blocks," said Peter K. Sculco, MD, orthopedic surgeon at HSS and also senior author on the study said. "We’re also looking at understanding prescribing patterns and how that affects how patients use their medication—if they are given a smaller prescription will they use less over time?"

The effort also has a strong patient-centered component. The new pain-management pathway includes educational programs for patients undergoing joint surgeries, including new additions to the preoperative education to manage pain expectations, early intervention by the pain management team for patients on preoperative or chronic opioids, and a multi-disciplinary inpatient team to treat and educate patients on alternatives to opioids for pain management. Although still in its early phase, data so far show that the pathway has led to a significant decrease in the use of opioids with no sacrifice in pain control. 

Dr. Sculco and his colleagues presented their findings at the 2019 annual meeting of the American Academy of Orthopaedic Surgeons in Las Vegas (exhibit SE35).

Other authors include: Ameer Elbuluk, MD; Michael Alexiades, MD; and Michael Cross, MD; all from HSS.

For more information on opioid prescription and pain management at HSS, please view the February 2019 issue of HSS Journal.

"Although concern remains regarding the risk of prothrombotic adverse events, including deep vein thrombosis, MI, pulmonary embolism and cerebrovascular events with routine use of TXA, the literature has continued to show that use of TXA does not increase the risk of thromboembolic events in TJA," he adds.

Read the full article at Healio.com/Orthopedics.
 

"The standards that we have used for diagnosing PJI in total knee replacement don’t necessarily apply to partial knee replacement. If you use the standard for a total knee replacement, then potentially you could be revising knee replacements for knees that are not actually infected," said principal study investigator Michael Cross, MD, an orthopedic surgeon at Hospital for Special Surgery, in New York City. "I think the guidelines need to be changed, so that partial knee replacement is a separate entity."

The International Consensus on Periprosthetic Joint Infection has established diagnostic criteria for diagnosing a PJI following total knee replacement, but cutoffs for diagnosis after a partial knee replacement remain unclear. "We have never really understood whether or not PJI should be diagnosed using the cutoffs for a total knee replacement," said Dr. Cross.

In the new study, Dr. Cross and colleagues reviewed records for all patients at HSS who had a partial knee replacement between August 2000 and December 2016 (n=234). They focused only on patients who had the complete data they were interested in analyzing, such as information on erythrocyte sedimentation rate (ESR), serum CO reactive protein (CRP), and synovial fluid white blood cell count (WBC). The researchers compared lab values for patients who were being revised for infection (n=18) to those who were being revised for reasons other than infection (n=76), to determine where the appropriate cutoffs were for diagnosing a PJI infection in the setting of a partial knee replacement.

The mean ESR (52.1 vs 17.2; P<0.01), CRP (106.4 vs 11.9; P<0.01), WBC (49,448 vs 967; P<0.01), and synovial PMNs (85.3 vs 47.3; P<0.01) were significantly higher in the infected group. The researchers identified optimal cutoff values for PJI in a multivariate model (ESR= 25.2 mm/h, CRP=17.2 mg/L, synovial WBC=6592.7, and %PMNs=71.7%). The synovial fluid WBC were higher than diagnostic cutoffs for total knee replacement.

"We looked at what is called the aspiration results, which is the amount of white blood cells and neutrophils in the joint fluid prior to revision. It is often one of the best tests we have," said Dr. Cross. "To make the diagnosis of infection in a partial knee replacement, the white blood cell cutoff was higher than in a total knee replacement. The cutoffs for diagnosis of infection in a partial knee replacement are somewhere in between a normal knee and a total knee replacement."

Dr. Cross said the findings are important for people trying to accurately treat patients with a partial knee replacement. "Just because the levels are slightly above the normal cutoffs for total knee replacement doesn’t necessarily mean that the knee is infected," said Dr. Cross.

PJI accounts for between 1.9 percent and 5 percent of revision procedures following partial knee arthroplasty. Roughly 10 percent of all knee replacement patients are a candidate for a partial knee replacement.