One of the sickest patients I’ve ever cared for was relatively well just a few days before he came to the hospital. This patient presented with septic arthritis of his knee, I performed an arthroscopic lavage, and it went rapidly downhill from there. His stormy course included multiple-system organ failure, a protracted stay in the intensive-care unit, and, eventually, death. This was early in my career, and I was heartsick. I conceived of this as a death after knee arthroscopy, and I had never heard of such a thing.
So I looked it up. It turns out that serious complications or death after septic arthritis occur commonly, particularly among patients who, like the patient I cared for, are older or have serious medical comorbidities [1, 3, 5]. One study found that 22.7% of patients with pyarthrosis of large joints will die, and nearly one in three will recover poorly with severe, permanent functional deficits ; another found the risk of death to be 9.3% within 30 days of diagnosis . In this month’s Clinical Orthopaedics and Related Research®, we learn that a history of intravenous drug use (a problem my patient also had) increases a patient’s odds of death from pyarthrosis nearly threefold, even after controlling for a number of important potential confounding variables .
This report by Dr. Matthew Salzler’s team at Tufts University School of Medicine in Boston also found that during the 14-year span of their study, there has been a shocking increase—a near-doubling—in the proportion of all hospitalizations that are related to complications from intravenous drug use; that figure now stands at 7%. The likelihood that a patient presenting with septic arthritis will have a history of intravenous drug use likewise more than doubled over that same period, to 11%. Even in the face of the now-well-known opioid epidemic, these figures are staggering.
On a tangential but important note about our specialty and its values, in the course of reviewing research related this month’s spotlight paper , I couldn’t help but notice how many of those papers begin by describing costs of care and problems imposed on the healthcare system [7, 9, 10]. I confess that I find this troubling, because this emphasis is mistaken. While these patients no doubt are expensive to care for—not least because as Oh and colleagues  determined, these patients are more likely to be under- or uninsured—that certainly is not the most-defining factor of this group of patients.
Our focus when caring for these patients (or when studying them) should be the same as with any other patients—on the individuals themselves, rather than the costs associated with caring for them. These patients should be important to us first because each of their experiences represents a sad, sad story. They often are the least among us; many of these patients live in dire poverty, with little or no social support, and with few available means of self-rescue. You may also have noticed that some providers who care for these patients focus on the volitional element—the bad choices that some of these patients have made to contribute to the problems from which they suffer. And undoubtedly volition plays a role, but if one takes the time to listen to these patients, quite often one finds that their own decisions often were not the etiologic or even a main determinative factor in causing these patients to be homeless, to have lost their families, to lose easy access to supportive social services and good preventative care, or to use drugs.
We need to force ourselves to listen to those stories. Caring for these patients effectively calls for a special kind of empathy.
Caring for these patients also calls for good data, including identification of factors associated with complications, precise estimates of the magnitudes of the risks involved, and the prevalence of the various endpoints of interest. For that reason, I especially appreciated this month’s excellent Editor’s Spotlight paper by Oh and colleagues . Their findings will help us to care for the specific medical and surgical concerns of these vulnerable patients more effectively. Please join me in the Take 5 interview that follows, as we take on the pressing social issues and other thought-provoking topics raised by this paper with its senior author, Matthew Salzler MD.
Take Five Interview with Matthew Salzler, senior author of “Increased Mortality and Reoperation Rates After Treatment for Septic Arthritis of the Knee in People Who Inject Drugs: Nationwide Inpatient Sample, 2000-2013”
Seth S. Leopold MD: Congratulations on this excellent work. How have you changed your approach to patients with septic arthritis as a result of your findings? What specifically do you do in terms of screening, débridement, and redébridement, and upon what do you base those changes?
Matthew Salzler MD: I learned a lot during the preparation of this manuscript and it caused me to examine how I deal with this patient population. The first change is simply to be more aware of how often patients who inject drugs present with septic arthritis of the knee. Oftentimes, we are consulted for “knee pain” in a patient with sepsis or bacterial endocarditis who is currently on antibiotics. The physical examination may be different or misleading in this setting, since the septic arthritis has been partially treated. I have a low threshold to aspirate the joint in these patients. Regarding the aspirate, the typical cell-count thresholds often do not apply to partially treated septic knees, and, in patients with ambiguous results, I have a discussion with our Infectious Disease team and the patient about the risks and benefits of repeated aspirations (while awaiting cultures) versus proceeding directly to surgery.
During surgery for septic knees, I consider the Gachter staging system , and vary my débridement accordingly. Oftentimes, patients who inject drugs have a delayed presentation and are more likely to show signs of chronic infection (Grade II or III) such as severe inflammation, synovial hypertrophy, and intrasynovial abscesses; even when they present with minimal purulence and synovial inflammation (Grade I), I now consider a more-aggressive synovectomy. The biggest change in management comes after surgery, as these patients are more likely to undergo a second procedure. If their postoperative pain, function, inflammatory markers, and drain output are not progressing as expected, I have a lower threshold for a return to the operating room.
Dr. Leopold: Can you give the reader a sense for the size and scope of this problem? Is this of subspecialty interest only, or does it touch all call-taking orthopaedic surgeons?
Dr. Salzler: The problem is larger than I wish it were. The estimate of persons who inject drugs is the United States is 2.6%  and the biggest limitation of this manuscript is that it likely underestimates the scope of the problem for two reasons. Drug use is often underdiagnosed, and, therefore, our algorithm to predict it is also likely an underestimate. Further, we only included patients with a principal diagnosis of septic arthritis of the knee; many patients who inject drugs present to hospitals with separate primary diagnoses and have septic arthritis of the knee as a secondary diagnosis. I would argue that it not only affects all call-taking orthopaedic surgeons, it also affects the multidisciplinary team of providers involved in the care of these patients.
Dr. Leopold: I recognize that our overarching goal in caring for patients is to practice evidence-based medicine. But I am struck when caring for patients with challenging social circumstances that the practice of narrative medicine may also be relevant . How can we use narrative medicine to deepen our abilities to empathize when caring for a patient with a history of intravenous drug use who presents with knee sepsis, and how might doing so change the patient’s circumstances over the longer run?
Dr. Salzler: The concept of narrative medicine has a role in both the patient’s and the provider’s experience in these situations. Many of us, myself included, chose our field because we like the instant gratification of successful surgical interventions with compliant patients and good clinical results. Treating patients who have drug addiction not only adds difficulties and complexities that come with the condition, but these patients also do not fare as well with the interventions we perform. It’s easy to dismiss these patients and become emotionally calloused to protect our egos. I certainly struggle with that as much as anyone else. However, awareness of this struggle can help me be engaged and empathetic in tough situations. Though I would love to say this means I am always successful in connecting with patients and changing their behavior, I fall short more often than I succeed. When that happens, I find solace in two areas. First, I am fortunate enough to work with an amazing multidisciplinary group at Tufts. These providers, such as my coauthor Dr. Alysse Wurcel of our Infectious Disease Department, deal with patients with complex social circumstances on a daily basis and consistently model exceptional and empathetic care. Second, in these situations, we need to remember that patient outcomes are not solely defined by a score on the SF-36 or KOOS, but they can be defined by the effort and care we provide.
Dr. Leopold: The 14-year story your paper tells is one of trendlines that all are moving in the wrong direction. Can an individual surgeon make a difference here, or is the better play to get involved with systems-based solutions? What looks most promising to you?
Dr. Salzler: This reminds me of the story that I have heard numerous times (adapted from The Star Thrower by Loren Eiseley); a boy walks along a beach on which thousands of starfish have washed up, and as he walks, he tosses them back into the ocean. An old man asks him why he bothers to do this, since there are far too many starfish on the shore for the child’s efforts to make a difference. The boy throws another one back, and answers, “it made a difference to that one!” That story aside, I think it’s important to focus on both individual and systems-based solutions. At Tufts, we treat these patients with a collaborative approach including social workers, addiction specialists, and infectious disease specialists. Though we hope to make a difference to our patients, we are aware that systems-based approaches are likely to have a broader effect, as my coauthor David Tybor, from our School of Public Health and Community Medicine, would agree. A public-health solution also has the benefit of focusing on prevention, which is necessary to address the growing opioid epidemic.
Dr. Leopold: Your study found associations with several sociodemographic factors that likely are related to poverty. To what degree are the problems you observed universal, or are they mainly phenomena of big-city hospitals and safety-net institutions?
Dr. Salzler: I wish I could say that these are isolated to cities and safety-net institutions, but addiction, substance use, and the downstream effects of these disorders are universal. Though poverty and substance use are correlated, it is likely more of a chicken and egg situation than a clear one-way causal relationship. Here, at Tufts Medical Center, we see septic arthritis in people who inject drugs who began their lives in nearly every socioeconomic and demographic background. Addiction can affect anyone.
1. Andreasen RA, Andersen NS, Just SA, Christensen R, Hansen I. Prognostic factors associated with mortality in patients with septic arthritis: A descriptive study. Scand J Rheumatol. 2017;46:27–32.
2. Charon R. Narrative medicine: A model for empathy, reflection, profession, and trust. JAMA. 2001;286:1897–1902.
3. Ferrand J, El Samad Y, Brunschweiler B, Grados F, Dehamchia-Rehailia N, Séjourne A, Schmit J-L, Gabrion A, Fardellone P, Paccou J. Morbimortality in adult patients with septic arthritis: A three-year hospital-based study. BMC Infect Dis. 2016;16:239.
4. Lansky A, Finlayson T, Johnson C, Holtzman D, Wejnert C, Mitsch A, Gust D, Chen R, Mizuno Y, Crepaz N. Estimating the number of persons who inject drugs in the United States by meta-analysis to calculate national rates of HIV and hepatitis C virus infections. PLoS One. 2014;9:e97596.
5. Maneiro JR, Souto A, Cervantes EC, Mera A, Carmona L, Gomez-Reino JJ. Predictors of treatment failure and mortality in native septic arthritis. Clin Rheumatol. 2015;34:1961–1967.
6. Oh DHW, Wurcel AG, Tybor DJ, Burke D, Menendez ME, Salzler MJ. Increased mortality and reoperation rates after treatment for septic arthritis of the knee in people who inject drugs: Nationwide inpatient sample, 2000-2013. Clin Orthop Relat Res. [Published online ahead of print]. DOI: 10.1007/s11999.0000000000000279.
7. Seymour RB, Ring D, Higgins T, Hsu JR. Leading the way to solutions to the opioid epidemic: AOA critical issues. J Bone Joint Surg. 2017;99:e113.
8. Stutz G, Guster MS, Kleinstück F, Gächter A. Arthroscopic management of septic arthritis: Stages of infection and results. Knee Surg Sports Traumatol Arthrosc. 2000;8:270–274.
9. Wang Z, Lenehan B, Itshayek E, Boyd M, Dvorak M, Fisher C, Kwon B, Paquette S, Street J. Primary pyogenic infection of the spine in intravenous drug users: A prospective observational study. Spine (Phila Pa 1976). 2012;15:685–692.
10. Ziu M, Dengler B, Cordell B, Bartanusz V. Diagnosis and management of primary pyogenic spinal infections in intravenous recreational drug users. Neurosurg Focus. 2014;37:E3