Periprosthetic joint infection (PJI) impacts a minority of total hip and total knee replacements but can impose substantial physical, emotional, and financial burdens on these patients and the health-care organizations engaging in their care. Alternative payment models have been proposed that will shift the direct cost of PJI treatment to hospital organizations and may impose financial penalties on health-care systems with PJI rates above national or regional averages. Kurtz et al. projected that the cost of managing PJI in the United States may exceed $1 billion by the year 20201.
Traditional approaches directed toward reducing PJI have focused on standardizing organizational protocols for the surgical procedure: hair removal, skin preparation, antibiotic selection, and antibiotic delivery2-4. Over the past decade, additional practices have been introduced to decrease wound-related complications: preoperative skin decolonization, the use of risk-adjusted venous thromboembolism (VTE) prophylaxis and antifibrinolytic agents, and the reduction of blood transfusion rates5-10.
A growing body of literature has recognized the contribution of patient factors in the risk for PJI. Several of these are cited in the study by Everhart et al., including morbid obesity, diabetes, tobacco abuse, malnutrition, inflammatory arthropathy, malignancy, previous surgical site infection (SSI), and coagulopathy. While the infection risks related to patient demographic features, medical comorbidities, and socioeconomic status have been previously reported, few have provided surgeons with effective tools to define SSI or PJI risks. Bozic et al. reported their development of a risk calculator for Medicare patients undergoing hip replacement surgery11, but predictive models for other patients are not, to our knowledge, presently available.
Everhart et al. have developed and validated a 35-point scoring system to help predict SSI risk. This was accomplished using the experience of a single institution over a 12-year interval (2000 to 2011). From an initial list of >45 comorbidities, 12 were incorporated into the SSI risk-assessment model. Overall, this process may be valuable to help define SSI risk for specific groups of patients with health features similar to those included in the study. The authors also identified thresholds when applying this scoring system (6 points for a primary total joint arthroplasty patient and 9 points for a revision total joint arthroplasty patient) where the risk for infection may exceed the benefits of elective intervention.
One limitation of the study is that it may not have been adequately inclusive to allow the assessment of many higher-risk patients undergoing elective hip and knee replacement surgery. The model included 6 risk factors that are not intrinsic to elective joint replacement (staphylococcal sepsis, prior PJI, osteomyelitis or septic arthritis, traumatic fracture, pathologic fracture, and bone neoplasm), a limited group of 4 core medical diagnoses (COPD [chronic obstructive pulmonary disease], diabetes, inflammatory arthropathy, and morbid obesity), and 2 chemical exposures (insulin use and tobacco use). There were other medical conditions present among the patients included in the study—renal failure, malnutrition, chronic anemia, acute blood-loss anemia, hepatitis C, somatoform or personality disorder, and RSD (reflex sympathetic dystrophy) or pain syndrome—that were not incorporated into the scoring system. The unadjusted odds ratio for infection for each of these conditions was higher than that of 2 of the nonelective risk factors that were used in the scoring model—traumatic fracture (2 points) and pathologic fracture (2.5 points). Since the study occurred across a period predating the introduction of several risk-reduction initiatives (skin decolonization, VTE risk stratification, and blood conservation approaches), this predictive model may not be generalizable to all patients undergoing arthroplasty today.
Even with those limitations, this is a valuable work that adds to the present body of literature that has identified the importance of considering patient demographic features and medical comorbidities in assessment of risk. Future efforts will be beneficial to identify the degree to which patient characteristics may be modifiable in order to provide a more predictable postoperative course and limit PJI risk following total hip or total knee replacement.
↵* The author indicated that no external funding was received for any aspect of this work. The Disclosure of Potential Conflicts of Interest form is provided with the online version of the article.
- Copyright © 2016 by The Journal of Bone and Joint Surgery, Incorporated