Study identifies individual risk factors linked to infection and mortality

Although total hip arthroplasty (THA) is a highly successful surgical procedure that restores mobility in many patients and leads to a better quality of life, a small number of THA patients experience complications, such as life-threatening periprosthetic joint infections (PJI), or die within 90 days of surgery. Certain comorbidities may be associated with increased risk of PJI and perioperative mortality, according to the results of a study presented at the 2010 annual meeting of the American Association of Hip and Knee Surgeons by Daniel J. Berry, MD, chair in the department of orthopaedic surgey at the Mayo Clinic and AAOS first vice president.

“Data demonstrate that PJI occurs in approximately 1 percent to 2 percent of patients during the first 10 years after hip arthroplasty, and that the 90-day mortality rate after THA ranges from 0.5 percent to 1 percent,” said Dr. Berry. Many published studies evaluate levels of risk using the Charleston index and the American Society of Anesthesiologists score, which represent aggregate scores of comorbidities.

“What has been lacking—and what is so important for risk adjustment—is to understand the degree to which specific comorbidities or diagnoses increase the risk of infection and mortality, and that is what this study aimed to do,” he said. “Learning more about these baseline comorbidities may help surgeons and patients better quantify the risks of surgery.”

Looking for risk factors
Dr. Berry and his fellow researchers evaluated data in the National Medicare 5 Percent Sample on 83,011 patients who underwent primary THA between 1998 and 2007 to identify baseline patient comorbidities associated with increased risk of infection and 90-day mortality.

“Patients were followed longitudinally with an encrypted identifier over the first 10 years after primary THA,” explained Dr. Berry. “Importantly, the presence of the comorbidity was determined through observation during the year prior to hip arthroplasty, not after the arthroplasty occurred.”

Investigators used a Cox Regression Model to examine 30 comorbid conditions, controlling for age, sex, race, census region, public assistance, and all other baseline comorbidities. The degree of association of each condition with infection or mortality was ranked using the p value associated with the test of significance for the hazard ratio.

Results
Two thirds of all patients had hypertension (Table 1). Other prevalent comorbidities included ischemic heart disease, hypercholesterolemia, history of malignancy, and cardiac arrhythmia.

Researchers found that the most significant comorbidities associated with PJI were rheumatologic disease, preoperative anemia, coagulopathy, diabetes, and depression. The most significant comorbidities related to perioperative mortality following THA were congestive heart failure, metastatic cancer, renal disease, dementia, and psychoses (Tables 2 and 3).

“Rheumatologic disease was the comorbidity that was most strongly associated with PJI, with a hazard ratio of 1.7, meaning the risk of PJI developing in a patient with rheumatologic disease was 1.7 times greater than in other patients,” he said. “Obesity had a hazard ratio of 1.7 as well, and coagulopathy had a hazard ratio of 1.6, most likely because of the risk of hematoma.”

Dr. Berry emphasized that a preoperative diagnosis of metastatic cancer tripled—and that a diagnosis of congestive heart failure more than doubled—the patient’s risk of 90-day mortality.

“Our study does have limitations,” acknowledged Dr. Berry. “The accuracy of coding of comorbidities is variable, and may be more specific than comprehensive. Obesity was almost certainly undercoded. If a condition is coded as being a comorbidity, the patient almost certainly had it, but not every single comorbidity may be noted in the database.

“And, without question,” he added, “the accuracy of coding prosthetic joint infection is not perfect, but it’s fairly good.”

The study has several strengths, noted Dr. Berry, including the fact that researchers evaluated the risk of comorbidities separately instead of using composite scores.

“In addition, this study used a large and varied sample,” he said.

Drawing conclusions
“We believe that these data can allow surgeons and patients to better quantify the risks of surgery for individual patients,” said Dr. Berry, “and potentially mitigate the risk of surgery for some of these factors.

“We also believe that refinement of these and other data will enable us to stratify risk more precisely, leading to fairer and more accurate comparisons of complication rates across healthcare systems,” he added.

Coauthors for “Risk Factors of Periprosthetic Joint Infection and Perioperative Mortality Following Total Hip Arthroplasty in Medicare Patients” include lead author Kevin J. Bozic, MD, MBA; Edmund Lau, MS; Kevin Ong, PhD; Steven M. Kurtz, PhD; Harry E. Rubash, MD; and Thomas P. Vail, MD.

Disclosure information: Dr. Berry—DePuy; Dr. Bozic—OREF, AHRQ, NIH, United Health Care, BCBSA, Integrated Healthcare Association, Pacific Business Group on Health, and CMS (MEDCAC); Mr. Lau—Stryker, Kyphon, Inc., Amgen Co., Alcon Corp.; Dr. Ong—Stryker, Medtronic; Dr. Kurtz—Stryker, Zimmer, Biomet, Medtronic, Synthes, Invibio, Stelkast, Ticona, Active Implants; Dr. Rubash—Biomet, Zimmer; Dr. Vail—Pivot Medical, DePuy.

Jennie McKee is a staff writer for AAOS Now. She can be reached at mckee@aaos.org

Bottom Line

• Patients with comorbidities such as rheumatologic disease, obesity, coagulopathy, chronic anemia, and diabetes may be at an increased risk of periprosthetic joint infection following THA.
• THA patients with congestive heart failure, psychosis, dementia, and renal disease may have an increased risk of 90-day postoperative mortality.
• Understanding the comorbidities associated with infection and mortality after THA will help surgeons and patients better understand the risks involved with the procedure.