Log in to MyACC
Preoperative Stress Testing in Total Hip and Knee Arthroplasty
Quick Takes
- Cardiac stress testing is frequently obtained prior to noncardiac surgery, though the trends of its use and the effects of testing on cardiovascular outcomes are unclear.
- The frequency of preoperative stress testing has declined annually from 2006-2017.
- Among patients with ≥1 RCRI condition, there was no difference in outcomes between those who did and did not undergo stress testing. Surprisingly, among patients with no RCRI conditions, the complication rate was twice as high in patients who did undergo stress testing compared to those who did not.
Study Questions:
How has the frequency of preoperative cardiac stress testing changed from 2004-2017, and what are patient outcomes during this time period?
Methods:
Using MarketScan inpatient claims data from 2003-2017, patients undergoing elective total knee arthroplasty and total hip arthroplasty were identified according to International Classification of Diseases (ICD)-9 and ICD-10 codes. The primary outcome measure was the frequency of cardiac stress testing within 60 days of elective surgery. With the further use of outpatient claims data and Current Procedural Terminology (CPT) codes, all types of cardiac stress tests (echocardiographic, nuclear, magnetic resonance imaging [MRI], and electrocardiographic [ECG]) were identified. Secondary outcomes included diagnoses of myocardial infarction (MI) or cardiac arrest during the inpatient perioperative period. A Revised Cardiac Risk Index (RCRI) score was calculated by assigning 1 point for each of the following diagnoses: ischemic heart disease, cerebrovascular disease, insulin therapy for diabetes, and chronic kidney disease based on stage III or higher (creatinine values were not consistently available). Joinpoint regression analysis was done to evaluate yearly trends of the unadjusted preoperative stress testing frequency.
Results:
The study cohort selection process yielded 801,396 elective joint replacement surgeries (hip 27.9%, knee 72.1%) from 2004-2017. Median age was 62 years (interquartile range, 57-70) and 58.1% were women. Overall, 10.4% (83,307) patients underwent preoperative cardiac stress testing within 60 days of surgery. An RCRI score could be calculated for 86.3% (71,905) of these patients, of which approximately half (35,531 out of 71,905, or 49%) had no RCRI conditions. The percentage of patients with RCRI scores of 0 increased from 44.7% in 2004 to 52.6% in 2017 (p < 0.001).
The rate of stress tests initially increased by 0.65% per year from 2004-2006. A Joinpoint was found in the third quarter of 2006, however, after which preoperative stress testing decreased 0.71% per year. The most frequent modality was nuclear stress testing (84.1% of patients), followed by stress echocardiography (11.3%) and MRI or ECG only (4.6%). All testing modalities decreased from 2006-2017, with nuclear stress testing decreasing the most. The total number of patients with an RCRI score of 0 undergoing stress testing was 6.7% (35,351 out of 527,047) compared to 23% (36,554 out of 159,020) for those with an RCRI score ≥1. The overall unadjusted complication rate for MI and cardiac arrest was 0.24% (n = 1,677 of 686,067). Rates of complications for patients with an RCRI score of 0 were 0.144% versus 0.58% for those with an RCRI score of 1. Complication rates decreased for both groups throughout the study period (p = 0.005). Among patients with an RCRI score of 0, those who underwent stress testing had a complication rate roughly twice that of patients who did not undergo stress testing (0.27% vs. 0.136%; p < 0.001). In contrast, there was no difference in complication rates among patients with an RCRI score ≥1 who underwent preoperative stress testing versus those who did not (0.60% vs. 0.57%; p = 0.51).
Conclusions:
Following an initial increase in stress testing with a peak frequency in 2006, there was a sustained decline in preoperative stress testing for patients undergoing total hip and knee replacements through 2017. Among patients with ≥1 RCRI condition, stress testing was done in 23% of patients. There was no difference in outcomes (MI or cardiac arrest) between those who did and did not undergo stress testing (0.60% vs. 0.57%). Among patients with no RCRI conditions, stress testing was done in 6.7% of patients. The complication rate was roughly twice as high in patients who did undergo stress testing compared to those who did not (0.27% vs. 0.136%).
Perspective:
This study showed a steady decline in preoperative stress testing from 2006 on. Despite this, the proportion of stress tests conducted in patients with RCRI scores of 0 remained relatively high (6.7%) despite newer American College of Cardiology/American Heart Association preoperative guidelines specifically discouraging this. On the one hand, this highlights somewhat limited adherence to guidelines and suggests an opportunity for education to further reduce nonindicated preoperative stress testing. On the other hand, it is surprising that the complication rate of patients with an RCRI score of 0 who underwent stress testing was twice as high as those who did not undergo testing. This suggests that risk stratification with the RCRI alone may not fully capture the risk of cardiac complications following knee and hip arthroplasty surgery. As outlined by the authors, it is also possible that the use of the MarketScan database underestimated patient comorbidities due to lack of coding for certain RCRI conditions. Further research is warranted to identify and evaluate the optimal patient conditions that warrant testing, and additional investigation will be needed to evaluate whether the findings from this study can be applied more broadly to other surgical procedures.
Clinical Topics: Arrhythmias and Clinical EP, Cardiovascular Care Team, Noninvasive Imaging, Prevention, Vascular Medicine, Implantable Devices, SCD/Ventricular Arrhythmias, Echocardiography/Ultrasound, Magnetic Resonance Imaging
Keywords: Arthroplasty, Replacement, Hip, Arthroplasty, Replacement, Knee, Cerebrovascular Disorders, Creatinine, Diabetes Mellitus, Diagnostic Imaging, Echocardiography, Stress, Electrocardiography, Exercise Test, Heart Arrest, Inpatients, Magnetic Resonance Imaging, Myocardial Infarction, Myocardial Ischemia, Outpatients, Perioperative Period, Preoperative Care, Renal Insufficiency, Chronic, Risk, Secondary Prevention
< Back to Listings
