Patterns of preoperative laboratory tests, EKG, and chest X-rays have not been previously described for the THR and TKR patient population. We aimed to describe these patterns in this patient population and found wide variation in practice, even among patients with no reported comorbidities. Moreover, while none of these tests was associated with length of stay, three tests (MRSA screening, urine culture, and EKG) were associated with lower odds of 90-day readmissions, and the reduction was 55–69%. These findings have important clinical and healthcare policy implications.
In the absence of clear criteria for which tests should be ordered for TKR and THR patients to evaluate them for surgery, we have shown wide variability in clinical practice. This variability has been reported in studies of other surgical specialties. For example, a national study analyzing preoperative testing for Medicare patients undergoing cataract surgery found nearly half of patients receive no testing before surgery. In contrast, others received at least one test . In another study of patients who underwent elective non-cardiac surgery, only 38% of patients underwent preoperative evaluation . Multiple factors have been cited in the literature that could explain this variation in the practice of preoperative testing. For example, medicolegal concerns, limited awareness of evidence-based guidelines, concerns about surgical cancellation, practice tradition, and beliefs about surgeons’ expectations may all play a role in ordering unnecessary tests . This variability could also be explained by variations among hospitals in the availability of workforce to universally establish preoperative screening clinics and among third-party payers in reimbursing for these tests [3, 22, 23]. On the policy level, bundled payment arrangements could induce this variation as not all arrangements cover preoperative testing . Further investigation is needed to understand the patient, hospital, geographic, and policy factors contributing to this variation.
Our study showed that three tests, the EKG, the MRSA screening, and the urine culture tests, were associated with substantially lower readmissions rates after surgery. These associations were robust, even in the subset of patients who did not have any reported comorbidities. These tests were among the least utilized in this patient population (only 11–13%). Our analysis showed a reduction in the odds of 90-day readmission by 69% in patients who underwent EKG screening. The American College of Cardiology/American Heart Association (ACC/AHA) guidelines recommend preoperative EKG for patients undergoing intermediate- and high-cardiac risk surgeries . Since the cardiac risk in orthopedic surgeries is intermediate , the findings of this study support the ACC/AHA recommendation. However, given the low rate of cardiac complications after total hip and knee replacements, we call for more granular studies to confirm our analysis.
EKG itself does not influence the cardiac risk, but it can serve as a baseline for comparison with the postoperative EKG if found abnormal . Adjustments in preoperative testing protocols to include EKG are relatively easy given that this test is low risk, cheap, and widely available. However, the low rate of EKG may be due to variations among internists who perform preoperative screening in categorizing different orthopedic procedures in the same bucket of intermediate cardiac risk . Further research is warranted to reveal the barriers to performing EKG for TKR and THR patients.
Our study also showed reduced readmissions among patients who had MRSA screening preoperatively. These results add to a growing body of evidence showing the utility of this test in preventing postoperative periprosthetic joint infection (PJI) among TKR and THR patients [12, 28]. Although our study evaluated all-cause readmissions, PJIs constitute about one-third of readmissions within the first few months [29, 30]. PJIs have significant adverse implications on patient outcomes and healthcare costs [31–33]. There remains a contentious debate in the orthopedic community regarding the utility of the MRSA test, which may explain the low utilization rate in our cohort . The American Academy of Orthopedic Surgeons (AAOS) has recently called for conducting a multicenter randomized controlled trial to evaluate the value of MRSA screening and decolonization in preventing PJI . While our cohort study was large and population-based, we support more rigorous definitive RCT evidence as suggested by the AAOS to address the conflicting views on this test.
Finally, our study showed that performing a urine culture before surgery was associated with lower odds of 90-days readmission. This test is usually utilized to screen for asymptomatic bacteriuria (ASB). Although screening for ASB is a longstanding practice , the limited and conflicting evidence on its benefit could explain the low utilization rate in our cohort . Two meta-analyses suggested an association between ASB and postoperative PJI after arthroplasty [37, 38]. They yet concluded that preoperative treatment of ASB is unnecessary due to the weak microbiological correlation between ASB and PJI  . It is important to note that most of the primary studies included in these meta-analyses are non-controlled small cohort studies and may have resulted in such a weak association. In contrast to the prior literature, our study is the first large population-based study, and it found a strong association between urine culture and 90-day readmission. Thus, this study may lay the ground for further investigation of the value of urine culture before TKR and THR.
There are implications to preoperative testing at the patient, provider, and insurer levels. For the patient, the impact of preoperative testing is not limited only to the direct costs of these tests but also involves the indirect costs in the form of travel and missed time from work to have these tests. They also involve additional stress and costs due to the false positives that could lead to further preoperative testing and precautions during and after the surgery. Moreover, these tests involve allocating the necessary resources (human and space) to hospitals to ensure patients are fit for surgery associated with high costs. However, identifying preoperative tests associated with lower readmission rates provides an opportunity for patient outcome improvement, and cost reduction is paramount, especially in the current environment of bundled payment schemes where hospitals shoulder readmission costs. The costs of these tests are minimal compared to the improved outcomes savings that could be realized from lower readmissions. As such, these results would be of use to all stakeholders.
This study has some limitations. First, our findings identify the tests associated with better outcomes among patients who underwent surgery. Since we do not observe in this dataset those who did not proceed forward with the surgery based on abnormal tests, we cannot assess the value of these tests. Therefore these findings should not belittle or negate the value of other lab tests in assessing preoperative risks . Second, this is a one-state study. While the state is large, geographically, and socio-demographically diverse, our findings may not represent the preoperative evaluation practices in all other states. Third, we excluded about half of the patients who underwent TKR and THR surgeries since they do not have associated preoperative screening codes. Our results might not generalize to the whole cohort because their characteristics differ from those with preoperative screening codes. Fourth, we possibly did not capture all the tests that patients underwent, as some tests could have been performed within a short period before the surgery (or for other reasons). Fifth, we did not perform a propensity score analysis as there were too many unobserved factors (both at patients’ and physicians’ levels) which would make matching ineffective. Finally, the results of these tests were not available in the SPARCS database, so we do not know whether the clinicians acted on these tests or not. Hence, we call for more granular clinical studies to confirm the results of this study, and we advocate against increasing the use of EKG, MRSA screening, and urine culture based solely on the results of this study.