NEW YORK (GenomeWeb) – Researchers at the Mayo Clinic are turning to metagenomic sequencing to identify pathogens responsible for infections that occur in hip and knee replacements.
During a presentation at the Individualizing Medicine Conference hosted by the Mayo Clinic last month, and in a follow-up interview with GenomeWeb, Robin Patel, director of the infectious diseases research laboratory at the Mayo Clinic, described a metagenomic sequencing approach her lab is testing to identify potential infections in patients whose joint replacements fail.
The technique is still too time consuming and expensive to be used clinically for diagnostic purposes, Patel said, but as technology improves and costs decline, she envisions eventually implementing a next-generation sequencing-based approach potentially for tricky cases of replacement failures.
Although the vast majority of hip and knee replacements are successful, the sheer number of the surgeries themselves have skyrocketed in the last decade and as the US population continues to age, are expected to continue to increase, Patel said. According to the US Centers for Disease Control and Prevention, in 2010, nearly 700,000 knee replacements and more than 310,000 hip replacements were performed in the US. By 2030, the American Academy of Orthopaedic Surgeons predicts that there will be nearly 3.5 million knee replacements and more than 500,000 hip replacements.
Patel said that these surgeries are typically extremely beneficial, enabling people who suffered from "limited mobility and lots of pain" to become active again, she said. But, approximately 1 percent of hip replacements and 2 percent of knee replacements become infected over a 10-year period. Infections develop in the biofilms on the surface of the implant, and do not result in typical symptoms associated with infection, she said. In addition, they can be slow to manifest, and therefore, slow to diagnose and intervene.
Successful management of an infected joint replacement typically involves one or two surgeries. One surgery to remove the infected join and to put in a so-called "spacer" coated with antibiotics. Patients also often get intravenous antibiotics, as well. Once the patient has recovered, the surgeon replaces the spacer with a new joint replacement.
Diagnosis can also be tricky, Patel added. "There are cases that look infected and we can't find the organism, or who don't look infected, but there is no explanation for why their replacement is failing," she said.
The current gold standard to diagnose microbial infections relies on culturing organisms extracted from the biofilm of the prosthetic. Patel's lab at the Mayo has come up with enhanced culturing techniques to help improve on diagnostic rates, but sensitivity of those approaches still hovers around 73 percent. The group also tested a PCR panel, which only boosted sensitivity to 77 percent.
In order to try and improve sensitivity as well as to identify the up to one-third of cases that have polymicrobial infections, her group decided to test metagenomic sequencing approaches.
The main challenge with metagenomic sequencing, Patel said, is to reduce the amount of background DNA, the majority of which is human DNA, and amplify the microbial DNA.
In a study published earlier this year in the Journal of Microbiological Methods, Patel's group tested New England Biolab's NEBNext Microbiome DNA enrichment kit and Molzym's MolYsis Basic kit.
The team is now also evaluating different approaches for whole-genome amplification to figure out which combination of technologies works best for their purposes. Patel noted in her presentation that the different kits have different advantages and biases.
In unpublished work, the group has analyzed samples from 332 patients whose joint replacement failed, including 162 who were not suspected to have an infection, 119 who were diagnosed with an infection, and 51 samples from patients thought to have an infection but whom a culture-based diagnostic turned up negative.
The metagenomic sequencing approach was concordant with 102 out of the 119 culture-positive samples. In six cases, the metagenomic approach did not detect a pathogenic infection and in 11 cases, it identified additional pathogens.
In addition, of the 51 samples that were culture negative but thought to be caused by infection, the metagenomic approach identified pathogens in 12 cases.
The metagenomic approach also enabled the team to assemble entire pathogen genomes, which enabled them to detect antibiotic-resistance genes, Patel said.
"This was a first look at the data," said Patel, whose team will continue to study the metagenomic approach under a grant from the National Institutes of Health.
"We're working through testing approximately 400 samples from patients who had a joint replacement, but had to have it removed because of infection or because it was failing" for another reason, she said.
She said the group is now "digging into the data" to look at the cases where metagenomic sequencing was discordant with the culture-based diagnostic to both figure out in the cases where the metagenomic sequencing approach did not identify a pathogen found in the culture and what was causing that discrepancy. Also, she said they will evaluate the cases that were culture negative but positive based on metagenomic sequencing in order to "convince ourselves of what is correct."
In addition, Patel said the team is continuing to refine their techniques to maximize microbial DNA content, minimize background DNA, and introduce as little bias as possible.
Currently, she said, the method is a great research tool to help understand these types of infections. For instance, she said, infections that develop within biofilms of joint replacements tend to be chronic and develop over a long period of time. "So, we're looking to see if these organisms have evolved into different subpopulations that could have implications in pathogenesis and treatment," she said.
Before it could be used for clinical diagnosis, the technology would have to significantly improve and the cost would need to come down, Patel said. However, technology such as Oxford Nanopore's MinIon could make such rapid and cheap metagenomic sequencing-based diagnostics possible, she said, noting that the platform still needed to improve significantly before it could accurately be used to diagnose infections.
She also anticipated that NGS-based diagnostics would first be used on special cases that were difficult to diagnose using standard methods.
"Hopefully, some of the work we're doing will define what's possible, and technology improvements will enable us to better bring this to the clinic," she said.
Patel said that her group is also planning to see whether they can use NGS approaches to diagnose infections noninvasively. Currently, to diagnose an infection of a hip or knee replacement, the joint itself has to be surgically removed in order to sample the biofilm on the surface of the joint. However, Patel's lab plans to assess whether sequencing of synovial fluid found in the joint could serve as a noninvasive means of diagnosing infection.
Other groups are also turning to metagenomic sequencing to diagnose unknown infections. A group at Cincinnati Children's Hospital has developed a metagenomic sequencing protocol for fecal samples to detect bacteria resistant to multiple drugs, and the University of California, San Francisco has been working on a metagenomic sequencing approaches on the MiSeq on the MinIon to diagnose infection.