Orthopedics

Feature Article 

Impact of Patent Foramen Ovale on Total Knee Arthroplasty Cerebrovascular Accident Perioperative Management

Cierra S. Hong, BA; Cary Politzer, MD; Sean P. Ryan, MD; Samuel S. Wellman, MD; William A. Jiranek, MD; Michael P. Bolognesi, MD; Thorsten M. Seyler, MD, PhD

Abstract

Venous thromboembolism and ischemic stroke are major complications following total knee arthroplasty (TKA) and potentially are associated with a patent foramen ovale (PFO). Although this association has been shown in other surgical disciplines, it has not been demonstrated in patients undergoing TKA. This study was undertaken to determine whether patients with a PFO would have a significantly increased risk of cerebrovascular accident (CVA) following TKA. The Humana national database was used to identify TKA patients who were stratified by the presence of a PFO from 2007 to 2016. Ninety-day follow-up was used for the primary outcome of CVA. Potential confounding comorbidities also were investigated, including age, sex, anticoagulation, insurance type, arrhythmia, valvular disease, peripheral vascular disease, chronic kidney disease, and diabetes mellitus. Of 153,245 TKAs, a total of 2272 patients had strokes; 479 of these patients had a PFO. On multivariable analysis, PFO remained an independent predictor of CVA postoperatively (odds ratio, 3.824; 95% confidence interval, 2.614–5.406; P<.0001). Other significant comorbidities associated with CVA included arrhythmia, chronic kidney disease, diabetes mellitus, peripheral vascular disease, and coronary valve disease. Importantly, low-molecular weight heparin or factor Xa inhibitor administration postoperatively had a negative correlation with stroke (odds ratio, 0.762; 95% confidence interval, 0.663–0.871; P=.0001 and odds ratio, 0.749; 95% confidence interval, 0.628–0.885; P=.0009, respectively). The findings of the multivariable analysis indicate PFO is associated with early postoperative CVA within 90 days following TKA. If known preoperatively, appropriate referral should be made to a cardiologist for PFO management and anticoagulation to reduce the overall risk of stroke. [Orthopedics. 2020;43(3):e151–e158.]

Abstract

Venous thromboembolism and ischemic stroke are major complications following total knee arthroplasty (TKA) and potentially are associated with a patent foramen ovale (PFO). Although this association has been shown in other surgical disciplines, it has not been demonstrated in patients undergoing TKA. This study was undertaken to determine whether patients with a PFO would have a significantly increased risk of cerebrovascular accident (CVA) following TKA. The Humana national database was used to identify TKA patients who were stratified by the presence of a PFO from 2007 to 2016. Ninety-day follow-up was used for the primary outcome of CVA. Potential confounding comorbidities also were investigated, including age, sex, anticoagulation, insurance type, arrhythmia, valvular disease, peripheral vascular disease, chronic kidney disease, and diabetes mellitus. Of 153,245 TKAs, a total of 2272 patients had strokes; 479 of these patients had a PFO. On multivariable analysis, PFO remained an independent predictor of CVA postoperatively (odds ratio, 3.824; 95% confidence interval, 2.614–5.406; P<.0001). Other significant comorbidities associated with CVA included arrhythmia, chronic kidney disease, diabetes mellitus, peripheral vascular disease, and coronary valve disease. Importantly, low-molecular weight heparin or factor Xa inhibitor administration postoperatively had a negative correlation with stroke (odds ratio, 0.762; 95% confidence interval, 0.663–0.871; P=.0001 and odds ratio, 0.749; 95% confidence interval, 0.628–0.885; P=.0009, respectively). The findings of the multivariable analysis indicate PFO is associated with early postoperative CVA within 90 days following TKA. If known preoperatively, appropriate referral should be made to a cardiologist for PFO management and anticoagulation to reduce the overall risk of stroke. [Orthopedics. 2020;43(3):e151–e158.]

Total knee arthroplasty (TKA) is one of the most common surgeries performed in the United States, with the projected number to reach almost 3.5 million procedures by 2030.1 Despite the increasing number of procedures, the rate of postoperative complications has declined significantly, related in part to advances in preoperative medical screening and management of medical comorbidities.2 In particular, ischemic stroke following elective surgery results in devastating outcomes for patients, and as such, efforts have been made to optimize at-risk patients.3 A more commonly discussed complication following elective arthroplasty is venous thromboembolism (VTE), which increases patients' risk for stroke4 and is treated prophylactically in all TKA patients. However, due to the limited evidence on the efficacy of various prophylactic protocols, every patient must be approached individually when considering anticoagulants and antiplatelet therapies.5

Clinically, the majority of patients with patent foramen ovale (PFO) are asymptomatic, but patient complaints can include shortness of breath, chest pains, and headaches secondary to conditions such as pulmonary embolisms, migraines, and obstructive sleep apnea, all of which are affected by right-to-left shunting.6 Patent foramen ovale also is associated with ischemic stroke and VTE within the general population.7,8 It is a preoperative risk factor that is under-addressed in the arthroplasty literature. The foramen ovale is a right-to-left atrial shunt present during fetal development to provide a path for oxygenated blood from the umbilical arteries to the fetus' body. Failure of the foramen ovale to close at birth (by fusing of the septum primum and secundum) creates a PFO, which occurs in approximately one-fourth of the general population and is mostly asymptomatic.6 This creates an open shunt that has been theorized to increase the risk of strokes in this patient population.9 Patent foramen ovale also has been shown to cause paradoxical emboli in the presence of VTE.10

Although PFO has been shown to increase the risk for stroke and VTE in patients undergoing cardiac11 and non-cardiac surgeries,12 it also has been associated with an increased risk for stroke in total hip arthroplasty (THA).13 However, to the authors' knowledge, the association between PFO and cerebrovascular accident (CVA) in TKA patients has not been evaluated previously. In this study, the authors sought to investigate the prevalence of PFO in TKA patients and to determine its relationship with CVA during the 90-day postoperative period. The authors hypothesized that patients with PFO would have a significantly increased risk of acute CVA after primary TKA.

Materials and Methods

The PearlDiver Research Program (PearlDiver, Warsaw, Indiana) is a national database that houses longitudinal medical data for millions of private payer patients.14 The authors used this program to search the Humana Inc database from 2007 to 2016 for primary TKA patients using Current Procedural Terminology (CPT) codes and International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9) procedural and International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10) procedural codes. Within this search, no patients younger than 40 years had a documented stroke within 90 days of surgery and therefore were excluded from analysis, resulting in a final cohort of 153,245 primary TKA patients. Using relevant ICD-9 and ICD-10 codes, patients with early CVA within 90 days of their surgery were identified and investigated within this cohort. Demographic information and medical comorbidities including age, sex, federal or private insurance, PFO, arrhythmia, valvular heart disease, peripheral vascular disease, chronic kidney disease, diabetes mellitus type 2, and various anticoagulation such as aspirin, fondaparinux, heparin, low-molecular weight heparin, warfarin, thrombin inhibitor, and factor Xa inhibitor also were included.

Statistical analysis was performed using the statistical software R (RStudio Inc, Boston, Massachusetts). For the primary outcome measure of CVA, both univariable and multivariable logistic regression analyses were performed to adjust for confounding variables. Data were presented as count (percentage) for all categorical variables including age, which was grouped as 40–44, 45–49, 50–54, 55–59, 60–64, 65–69, 70–74, 75–79, 80–84, 85–89, and older than 90 years. On univariable analysis, risk factors correlated with CVA (P<.25) were included in a multivariable logistic regression analysis to determine the relative importance of a PFO. Data from the multivariable regression were presented as odds ratios (OR) with 95% confidence intervals (CI). A Bonferroni-corrected P for multiple comparisons of P<.0019 determined statistical significance. Queries that produced a cohort of 11 or fewer patients represented null values in the PearlDiver program to protect patient identification per the Health Insurance Portability and Accountability Act. However, this did not affect the analysis, which used patient-specific multivariate analysis. All collected data were deidentified and exempted from institutional review board approval at the authors' institution.

Results

Of 153,245 patients who underwent primary TKA from 2007 to 2016, a total of 2272 patients had a stroke within 90 days of their surgery. Significant differences in various age ranges, sex, insurance, medical comorbidities, and anticoagulation treatments were observed in patients who experienced early CVA vs patients who did not (Tables 12). Specifically, patients who had a CVA during their 90-day postoperative period were more likely to have cardiac arrhythmias, heart valvular disease, chronic kidney disease, diabetes mellitus, perivascular disease, and PFO.

Demographics of TKA Patients With and Without Stroke

Table 1:

Demographics of TKA Patients With and Without Stroke

Medical Comorbidities and Anticoagulation of Primary TKA Patients With and Without Stroke

Table 2:

Medical Comorbidities and Anticoagulation of Primary TKA Patients With and Without Stroke

Based on the independent predictors (P<.25) found in the univariable analysis of strokes, a multivariable logistic regression analysis was performed to determine the relative significance of PFO (Table 3). In patients who had a CVA within 90 days of their primary TKA, PFO was a significant independent predictor postoperatively for this complication (OR, 3.824; 95% CI, 2.614–5.406; P<.0001) with the greatest OR of all factors. Other significant independent predictors for postoperative CVA were arrhythmias, valvular disease, chronic kidney disease, diabetes mellitus, and peripheral vascular disease, which is consistent with previously reported literature.15–18 Patients who had postoperative CVAs with (479 patients) and without PFO (152,766 patients) then were stratified to assess demographics, medical comorbidities, and anticoagulation (Table 4).

Confounding Variables in Multivariate Regression Analysis in Primary TKA Patients With and Without CVA

Table 3:

Confounding Variables in Multivariate Regression Analysis in Primary TKA Patients With and Without CVA

Demographics, Medical Comorbidities, and Anticoagulation Therapy of Primary TKA Patients With and Without Patent Foramen Ovale

Table 4:

Demographics, Medical Comorbidities, and Anticoagulation Therapy of Primary TKA Patients With and Without Patent Foramen Ovale

Discussion

In the United States, approximately 800,000 individuals have a CVA every year and approximately 75% are first-time or new strokes. Cerebrovascular accident is one of the leading causes of deaths in the country, with the cost to patients and the health care system exceeding $300 billion annually.19 Perioperative strokes in particular are relatively uncommon, affecting less than 1% of patients undergoing non-cardiovascular surgeries,20 but these strokes have been widely studied due to the high morbidity and mortality.3,16,20 Most of the published literature assessing the impact of perioperative stroke in total joint arthroplasty focuses on patients undergoing THA. Bateman et al21 used the National Inpatient Sample and reported a CVA incidence of 0.2% to 0.3% in primary THA patients from 2000 to 2004. Their analysis showed that renal insufficiency, cardiac arrhythmias, and valvular disease had the highest impact on predicting the risk for perioperative CVA. In a Danish database from 1998 to 2007, Lalmohamed et al22 also found the risk for CVA, especially ischemic CVA, was elevated more than fourfold in the first 6 weeks after a patient's THA.

Rasouli et al23 acknowledged the gap in the total joint arthroplasty literature and evaluated perioperative CVA in primary and revision TKA and THA from 2002 to 2011. The overall incidence of stroke decreased from 0.17% to 0.14% during the 9-year study period, but the inpatient mortality was significantly higher for these patients (OR, 27.73; 95% CI, 23.06–33.05; P<.001). In contrast, a Danish study by Pedersen et al24 showed the risks of postoperative VTE and stroke within 90 days of primary THA and TKA were relatively unchanged at 1.3% and 0.5%, respectively, during a 15-year period despite appropriate thromboprophylaxis. In both of these studies, medical comorbidities including renal disease, cardiac arrhythmia, valvular disease, and diabetes mellitus type 2 remained the most significant risk factors for these incidences.

As previously stated, various medical diseases, specifically PFO, have been shown to increase the risk of CVA.25 Perfetti et al13 investigated the risk of perioperative ischemic strokes in almost 400,000 THA patients from 2007 to 2013 with atrial septal defects or PFO using the National Inpatient Sample and reported a 29-fold increased risk (95% CI, 6.68–125.89; P<.001) of acute CVA compared with matched controls (7.14% vs 0.26%; P<.001). In the TKA patient population, the literature assessing the preoperative risk of CVA in the presence of PFO is limited. There have been a few case reports of patients with underlying right-to-left shunts via PFO who develop early postoperative CVA following TKA,26,27 but no large database study exists. Therefore, this study not only contributes to the current knowledge of PFO but also is the first study using a national database to show that following primary TKA, PFO is a significant independent predictor of early postoperative CVA. In addition, in the current analysis, PFO was the most significant correlate compared with other medical comorbidities associated with CVA.

With this finding, the question of how best to manage these patients preoperatively becomes important. To assess for underlying cardiac structural causes such as right-to-left shunts, patients can be referred for transthoracic echocardiography with contrast or bubble study via imaging and various maneuvers such as Valsalva or coughs.28 The best imaging modality to diagnose PFO due to the location of the abnormality is transesophageal echocardiography; however, additional resources and requirements, such as patient sedation, are needed to perform this test.28,29 In addition, such tests can cost from $255 to more than $2000.30

If a PFO is found, the next question is whether to close the shunt. A recent meta-analysis for secondary prevention of cryptogenic ischemic stroke by Schulze et al31 evaluated the benefits of PFO closure compared with medical therapy from 5 randomized-controlled trials, only 1 of which had longitudinal follow-up results published. These trials were CLOSURE I,32 PC Trial,33 RESPECT,34 Gore REDUCE,35 CLOSE,36 and long-term endpoints of RESPECT.37 When the first 3 studies were published, there was debate regarding the outcomes due to the potential patient selection bias and variety in coexisting medical comorbidities and shunt sizes. However, with more longitudinal data, PFO closure overall had a significantly positive effect on reducing recurrent CVA compared with medical therapies alone, including antiplatelet and anticoagulant drugs, after accounting for the heterogeneity between the studies in the analysis.31 Despite this, there is no approved indication for PFO closure to prevent recurrent CVA.

For patients with a PFO but no history of CVA, it is difficult to assess the need for PFO closure, which has an increased risk of new-onset atrial fibrillation or flutter. Other less common complications include myocardial infarction, infective endocarditis, sepsis, and death.38 Due to the increased risk of CVA shown in the current study, if it is known that a patient has risk factors or the presence of a PFO prior to undergoing TKA, appropriate referrals and collaborations with the patient's cardiologist should be made for discussions regarding potential procedural intervention.

Another option is prophylactic anticoagulation to minimize the risk of CVA in this at-risk patient population. As discussed by Lieberman and Heckmann,5 the American College of Chest Physicians 2012 guidelines recommend low-molecular weight heparin, direct (ie, apixaban, rivaroxaban) or indirect (ie, fondaparinux) factor Xa inhibitors, vitamin K antagonists (ie, warfarin), aspirin, or mechanical compression for at least 10 to 14 days for VTE prophylaxis after THA and TKA. On the other hand, the American Academy of Orthopaedic Surgeons does not recommend specific treatment regimens or durations due to the complex interactions of these medications and a patient's medical comorbidities.

Different randomized controlled trials have shown extended aspirin VTE prophylaxis after THA or TKA had similar outcomes to rivaroxaban39 and dalteparin,40 a low-molecular weight heparin, but the relative efficacy of these drugs in patients with PFO is unknown. In this study, low-molecular weight heparin and factor Xa inhibitors were found to be protective of strokes in the general population after TKA (Table 2). Unfortunately, due to the limited size of the study's cohort, the authors were unable to assess whether these anticoagulants are also protective of strokes after TKA in patients with PFO compared with patients without PFO. Therefore, low-molecular weight heparin and factor Xa inhibitors require further study to determine their efficacy as CVA chemoprophylaxis for patients with PFO. Meanwhile, as stated previously, patients with concerns of a PFO should be referred to a cardiologist, who will be able to make the best-informed decision regarding patients' anticoagulation therapy.

There were some limitations to this study and its retrospective design. First, patient data were collected solely from the Humana database, which can lead to selection bias due to the private-payer population. However, this database was chosen over others such as the standard analytical files because not only was patient information available after 2014, but the authors also were able to obtain medication data. Patients who were younger than 40 years also were omitted because none of the patients in the database had documented strokes, and the authors were unable to collect data on patients with a PFO due to the small cohort size. This leads to some selection bias, but the authors believed that if this young cohort were included, which was not their population of interest as the majority of patients receiving TKA are older than 60 years, it could have confounded the results and made other comorbidities insignificant. In addition, the authors used this platform because of its capabilities to perform a multivariate regression analysis. Despite the lack of propensity patient matching in this study, the analysis provided an alternative way of controlling for confounding variables when analyzing patients with PFO to those without in relation to the incidence of stroke.

Another limitation was the use of billing codes, both ICD-9 and ICD-10, to determine clinical diagnoses, which can lead to coding inconsistencies and errors. In general, however, the validity of clinical diagnoses between the 2 systems are similar,41 especially regarding the diagnostic codes for acute strokes.42 Unfortunately, as reported in recent studies,43,44 the coding for PFO in both systems has errors regardless of the transition and is a general limitation in studies using billing codes. On the other hand, the coding for PFO used in this study was likely to accurately represent patients with a true atrial septal defect.43 Finally, the correlation between low-molecular weight heparin and factor Xa inhibitors and decreased stroke risk in patients with PFO may be limited as this study did not take into account the different routes of administration and dosage among patients.

Despite the rare occurrence of PFO, with TKA being one of the highest, if not the most common, surgery in the country in a potentially older and more medically complicated patient population, it is important to minimize all preoperative risk factors with medical optimization. As previously reported, the overall trend of perioperative CVA is decreasing; however, there is still a need to continue to improve management, as well as education, of these risks, especially to patients with PFO undergoing TKA. Further research is needed to explore the benefits of prophylactic PFO closure in patients who are CVA-naïve as well as to clarify the possible indications for PFO closure in patients who experience strokes secondary to their shunt. This would allow the clinical community to investigate the cost effectiveness of preoperatively screening all TKA patients for PFO to reduce the incidence of CVA and its associated mortality and morbidity. In addition, the advantage of various anticoagulation to protect patients with PFO against CVA needs to be assessed.

Conclusion

This study investigated the impact of PFO on the pre- and postoperative management of CVA in primary TKA patients. In addition to reporting similar risk factors associated with CVA after primary TKA as seen in various surgical procedures, an increased risk for early CVA was found in patients with PFO after primary TKA. As noted, the study had several limitations due to the type of database and the retrospective investigation, but the findings highlight a poorly addressed issue in the arthroplasty literature. By showing the relationship between acute postoperative CVA and PFO in patients receiving primary TKA, the authors have highlighted the lack of conclusive literature on the perioperative management of patients with PFO undergoing TKA and the need for further research to decrease CVA risk in this patient population.

References

  1. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007;89(4):780–785. PMID: 17403800
  2. Oviedo Baena AM, Moeschler SM, Smith HM, Duncan CM, Schroeder DR, Kopp SL. Perioperative comorbidities and complications among patients undergoing primary total knee arthroplasty: a retrospective analysis and prospective survey. J Clin Anesth. 2015;27(7):558–565. doi:10.1016/j.jclinane.2015.07.011 [CrossRef] PMID:26337563
  3. Amin H, Greer DM. Perioperative stroke. In: The Stroke Book. 2nd ed. Cambridge, UK: Cambridge University Press; 2013:257–268.
  4. Rinde LB, Småbrekke B, Mathiesen EB, et al. Ischemic stroke and risk of venous thromboembolism in the general population: the Tromsø study. J Am Heart Assoc. 2016;5(11):e004311. doi:10.1161/JAHA.116.004311 [CrossRef] PMID:27821402
  5. Lieberman JR, Heckmann N. Venous thromboembolism prophylaxis in total hip arthroplasty and total knee arthroplasty patients: from guidelines to practice. J Am Acad Orthop Surg. 2017;25(12):789–798. doi:10.5435/JAAOS-D-15-00760 [CrossRef] PMID:29176502
  6. Asrress KN, Marciniak M, Marciniak A, Rajani R, Clapp B. Patent foramen ovale: the current state of play. Heart. 2015;101(23):1916–1925. doi:10.1136/heartjnl-2015-307639 [CrossRef] PMID:26487104
  7. Homma S, Di Tullio MR. Patent foramen ovale and stroke. J Cardiol. 2010;56(2):134–141. doi:10.1016/j.jjcc.2010.05.008 [CrossRef] PMID:20591626
  8. Grander W, Schachner T, Velik-Salchner C, Dünser MW. Patent foramen ovale and major pulmonary embolism. J Cardiothorac Vasc Anesth. 2011;25(5):841–843. doi:10.1053/j.jvca.2010.06.024 [CrossRef] PMID:20817561
  9. Landzberg MJ, Khairy P. Patent foramen ovale: when is intervention warranted?Can J Cardiol. 2013;29(7):890–892. doi:10.1016/j.cjca.2013.04.019 [CrossRef] PMID:23790552
  10. Dalen JE, Alpert JS. Cryptogenic strokes and patent foramen ovales: what's the right treatment?Am J Med. 2016;129(11):1159–1162. doi:10.1016/j.amjmed.2016.08.006 [CrossRef] PMID:27566504
  11. Kutty S, Sengupta PP, Khandheria BK. Patent foramen ovale: the known and the to be known. J Am Coll Cardiol. 2012;59(19):1665–1671. doi:10.1016/j.jacc.2011.09.085 [CrossRef] PMID:22554596
  12. Ng PY, Ng AKY, Subramaniam B, et al. Association of preoperatively diagnosed patent foramen ovale with perioperative ischemic stroke. JAMA. 2018;319(5):452–462. doi:10.1001/jama.2017.21899 [CrossRef] PMID:29411032
  13. Perfetti DC, Chughtai M, Boylan MR, Naziri Q, Maheshwari AV, Mont MA. Atrial septal defect increases the risk for stroke after total hip arthroplasty. J Arthroplasty. 2017;32(10):3152–3156. doi:10.1016/j.arth.2017.05.006 [CrossRef] PMID:28602537
  14. PearlDiver. About PearlDiver. http://www.pearl-diverinc.com/about-us.php. Accessed August 27, 2018.
  15. Orenes-Piñero E, Montoro-García S, Banerjee A, Valdés M, Lip GYH, Marín F. Pre and post-operative treatments for prevention of atrial fibrillation after cardiac surgery. Mini Rev Med Chem. 2012;12(13):1419–1431. PMID: 23106292
  16. Ahmad S, Wilt H. Stroke prevention in atrial fibrillation and valvular heart disease. Open Cardiovasc Med J. 2016;10(suppl-1, M6):110–116. doi:10.2174/1874192401610010110 [CrossRef];
  17. Toyoda K, Ninomiya T. Stroke and cerebrovascular diseases in patients with chronic kidney disease. Lancet Neurol. 2014;13(8):823–833. doi:10.1016/S1474-4422(14)70026-2 [CrossRef] PMID:25030514
  18. Nyström T, Holzmann MJ, Sartipy U. Long-term risk of stroke in patients with type 1 and type 2 diabetes following coronary artery bypass grafting. J Am Heart Assoc. 2015;4(11):e002411. doi:10.1161/JAHA.115.002411 [CrossRef] PMID:26553216
  19. Benjamin EJ, Blaha MJ, Chiuve SE, et al. Heart disease and stroke statistics: 2017 update. A Report From the American Heart Association. Circulation. 2017;135(10):e146–e603.
  20. Kikura M, Bateman BT, Tanaka KA. Perioperative ischemic stroke in non-cardiovascular surgery patients. J Anesth. 2010;24(5):733–738. doi:10.1007/s00540-010-0969-3 [CrossRef] PMID:20549522
  21. Bateman BT, Schumacher HC, Wang S, Shaefi S, Berman MF. Perioperative acute ischemic stroke in noncardiac and nonvascular surgery: incidence, risk factors, and outcomes. Anesthesiology. 2009;110(2):231–238. doi:10.1097/ALN.0b013e318194b5ff [CrossRef] PMID:19194149
  22. Lalmohamed A, Vestergaard P, Cooper C, et al. Timing of stroke in patients undergoing total hip replacement and matched controls: a nationwide cohort study. Stroke.2012;43(12):3225–3229. doi:10.1161/STROKEAHA.112.668509 [CrossRef] PMID:23132782
  23. Rasouli MR, Tabatabaee RM, Maltenfort MG, Chen AF. Acute stroke after total joint arthroplasty: a population-based trend analysis. J Clin Anesth. 2016;34:15–20. doi:10.1016/j.jclinane.2016.03.034 [CrossRef] PMID:27687339
  24. Pedersen AB, Mehnert F, Sorensen HT, Emmeluth C, Overgaard S, Johnsen SP. The risk of venous thromboembolism, myocardial infarction, stroke, major bleeding and death in patients undergoing total hip and knee replacement: a 15-year retrospective cohort study of routine clinical practice. Bone Joint J. 2014;96B(4):479–485. doi:10.1302/0301-620X.96B4.33209 [CrossRef] PMID:24692614
  25. Lechat P, Mas JL, Lascault G, et al. Prevalence of patent foramen ovale in patients with stroke. N Engl J Med. 1988;318(18):1148–1152. doi:10.1056/NEJM198805053181802 [CrossRef] PMID:3362165
  26. Bruce-Brand R, Dragomir R, Baker J, Harty J. Cerebrovascular infarction following bilateral total knee arthroplasty and tranexamic acid administration. Acta Orthop Belg. 2013;79(3):351–354. PMID: 23926741
  27. Cho SY, Youn HJ, Park MY, et al. Post-operative multiple thrombosis associated with patent foramen ovale: embolic stroke, right atrial thrombi, pulmonary embolism and deep vein thrombosis. J Cardiovasc Ultrasound. 2015;23(3):177–180. doi:10.4250/jcu.2015.23.3.177 [CrossRef] PMID:26448826
  28. Rana BS, Thomas MR, Calvert PA, Monaghan MJ, Hildick-Smith D. Echocardiographic evaluation of patent foramen ovale prior to device closure. JACC Cardiovasc Imaging. 2010;3(7):749–760. doi:10.1016/j.jcmg.2010.01.007 [CrossRef] PMID:20633854
  29. Pinto FJ. When and how to diagnose patent foramen ovale. Heart. 2005;91(4):438–440. doi:10.1136/hrt.2004.052233 [CrossRef]
  30. Healthcare Bluebook. Retrieved from https://www.healthcarebluebook.com. Accessed July 28, 2019.
  31. Schulze V, Lin Y, Karathanos A, et al. Patent foramen ovale closure or medical therapy for cryptogenic ischemic stroke: an updated meta-analysis of randomized controlled trials. Clin Res Cardiol. 2018;107(9):745–755. doi:10.1007/s00392-018-1224-4 [CrossRef] PMID:29500568
  32. Furlan AJ, Reisman M, Massaro J, et al. CLOSURE I Investigators. Closure or medical therapy for cryptogenic stroke with patent foramen ovale. N Engl J Med. 2012;366(11):991–999. doi:10.1056/NEJMoa1009639 [CrossRef] PMID:22417252
  33. Meier B, Kalesan B, Mattle HP, et al. PC Trial Investigators. Percutaneous closure of patent foramen ovale in cryptogenic embolism. N Engl J Med. 2013;368(12):1083–1091. doi:10.1056/NEJMoa1211716 [CrossRef] PMID:23514285
  34. Carroll JD, Saver JL, Thaler DE, et al. RESPECT Investigators. Closure of patent foramen ovale versus medical therapy after cryptogenic stroke. N Engl J Med. 2013;368(12):1092–1100. doi:10.1056/NEJMoa1301440 [CrossRef] PMID:23514286
  35. Søndergaard L, Kasner SE, Rhodes JF, et al. Gore REDUCE Clinical Study Investigators. Patent foramen ovale closure or antiplatelet therapy for cryptogenic stroke. N Engl J Med. 2017;377(11):1033–1042. doi:10.1056/NEJMoa1707404 [CrossRef] PMID:28902580
  36. Mas J-L, Derumeaux G, Guillon B, et al. CLOSE Investigators. Patent foramen ovale closure or anticoagulation vs antiplatelets after stroke. N Engl J Med. 2017;377(11):1011–1021. doi:10.1056/NEJMoa1705915 [CrossRef] PMID:28902593
  37. Saver JL, Carroll JD, Thaler DE, et al. RESPECT Investigators. Long-term outcomes of patent foramen ovale closure or medical therapy after stroke. N Engl J Med. 2017;377(11):1022–1032. doi:10.1056/NEJMoa1610057 [CrossRef] PMID:28902590
  38. Merkler AE, Gialdini G, Yaghi S, et al. Long-term risk of complications after endovascular closure of patent foramen ovale. Stroke Conf Am Hear Assoc Stroke Assoc. 2017;48(suppl. 1).
  39. Anderson DR, Dunbar M, Murnaghan J, et al. Aspirin or rivaroxaban for VTE prophylaxis after hip or knee arthroplasty. N Engl J Med. 2018;378(8):699–707. doi:10.1056/NEJMoa1712746 [CrossRef] PMID:29466159
  40. Anderson DR, Dunbar MJ, Bohm ER, et al. Aspirin versus low-molecular-weight heparin for extended venous thromboembolism prophylaxis after total hip arthroplasty: a randomized trial. Ann Intern Med. 2013;158(11):800–806. doi:10.7326/0003-4819-158-11-201306040-00004 [CrossRef] PMID:23732713
  41. Quan H, Li B, Saunders LD, et al. IMECCHI Investigators. Assessing validity of ICD-9-CM and ICD-10 administrative data in recording clinical conditions in a unique dually coded database. Health Serv Res. 2008;43(4):1424–1441. doi:10.1111/j.1475-6773.2007.00822.x [CrossRef] PMID:18756617
  42. McCormick N, Bhole V, Lacaille D, Avina-Zubieta JA. Validity of diagnostic codes for acute stroke in administrative databases: a systematic review. PLoS One. 2015;10(8):e0135834. doi:10.1371/journal.pone.0135834 [CrossRef] PMID:26292280
  43. Rodriguez FH III, Ephrem G, Gerardin JF, Raskind-Hood C, Hogue C, Book W. The 745.5 issue in code-based, adult congenital heart disease population studies: relevance to current and future ICD-9-CM and ICD-10-CM studies. Congenit Heart Dis. 2018;13(1):59–64. doi:10.1111/chd.12563 [CrossRef] PMID:29266726
  44. Khan A, Ramsey K, Ballard C, et al. Limited accuracy of administrative data for the identification and classification of adult congenital heart disease. J Am Heart Assoc. 2018;7(2):e007378. doi:10.1161/JAHA.117.007378 [CrossRef] PMID:29330259

Demographics of TKA Patients With and Without Stroke

CharacteristicNo.P

Stroke (N=2272)No Stroke (N=150,973)
Age, y
  40–44a631 (0.4%)b
  45–4912 (0.5%)1951 (1.3%).0019
  50–5452 (2.3%)5284 (3.5%).0020
  55–59112 (4.9%)10,288 (6.8%).0004
  60–64187 (8.2%)15,461 (10.2%).0018
  65–69439 (19.3%)37,680 (25.0%)<.0001
  70–74541 (23.8%)36,761 (24.4%).5648
  75–79471 (20.7%)25,362 (16.8%)<.0001
  80–84287 (12.6%)12,513 (8.3%)<.0001
  85–8974 (3.3%)2770 (1.8%)<.0001
  >9090 (4.0%)2272 (1.5%)<.0001
Sex (female)1342 (59.1%)95,105 (63.0%).0001
Insurance
  Fully insured with Humana89 (3.9%)16,525 (11.0%)<.0001
  Medicare2182 (96.0%)134,248 (88.8%)<.0001
  Medicaida459 (0.3%)b

Medical Comorbidities and Anticoagulation of Primary TKA Patients With and Without Stroke

CharacteristicStroke (N=2272)No Stroke (N=150,973)P
Medical comorbidity, No.
  Arrhythmia1217 (53.6%)53,358 (35.3%)<.0001
  Valvular disease1071 (47.1%)43,052 (28.5%)<.0001
  Chronic kidney disease935 (41.2%)39,963 (26.4%)<.0001
  Diabetes mellitus type 21321 (58.1%)66,503 (44.0%)<.0001
  Perivascular disease799 (35.2%)29,858 (19.8%)<.0001
  Patent foramen ovale33 (1.45%)446 (0.3%)<.0001
Anticoagulation, No.
  Aspirin18 (0.8%)1837 (1.2%).0680
  Fondaparinux20 (0.9%)2036 (1.4%).0556
  Heparin0 (0%)28 (0%)a
  Low-molecular weight heparin252 (11.1%)20,955 (13.9%).0002
  Warfarin490 (21.6%)30,649 (20.3%).1369
  Thrombin inhibitorb175 (0.1%)a
  Factor Xa inhibitor154 (6.8%)14,336 (9.5%)<.0001
Length of stay, median (lower quartile, upper quartile), d3 (0, 5)3 (0, 3)<.0001

Confounding Variables in Multivariate Regression Analysis in Primary TKA Patients With and Without CVA

ComorbidityOdds Ratio95% Confidence IntervalPa
Age, y
  45–490.5350.213–1.446.1913
  50–540.7690.370–1.870.5191
  55–590.7620.379–1.817.4905
  60–640.7530.379–1.781.4650
  65–690.5990.304–1.410.1836
  70–740.6740.342–1.586.3055
  75–790.7630.386–1.797.4829
  80–840.8640.436–2.043.7070
  85–890.9820.478–2.372.9637
  >901.3170.646–3.167.4902
Sex (male)1.1191.026–1.220.0109
Medical comorbidity
  Arrhythmia1.4581.330–1.598<.0001
  Valvular disease1.5071.374–1.653<.0001
  Chronic kidney disease1.2581.146–1.380<.0001
  Diabetes mellitus1.4021.282–1.534<.0001
  Peripheral vascular disease1.4791.345–1.625<.0001
  Patent foramen ovale3.8242.614–5.406<.0001
Anticoagulation
  Aspirin0.9450.569–1.466.8138
  Fondaparinux0.6810.422–1.032.0899
  Low-molecular weight heparin0.7620.663–0.871.0001
  Warfarin0.9740.876–1.082.6294
  Thrombin inhibitor1.2490.441–2.755.6263
  Factor Xa inhibitor0.7490.628–0.885.0009
Insurance
  Fully insured with Humana0.4640.363–0.586<.0001
  Medicare0.6730.207–1.585.4333
Length of stay1.0231.016–1.031<.0001

Demographics, Medical Comorbidities, and Anticoagulation Therapy of Primary TKA Patients With and Without Patent Foramen Ovale

CharacteristicPatent Foramen Ovale (N=479)No Patent Foramen Ovale (N=152,766)P
Age, y, No.
  40–44a637 (0.4%)b
  45–49a1958 (1.3%)b
  50–5417 (3.5%)5319 (3.5%).9416
  55–5921 (4.4%)10,379 (6.8%).0004
  60–6444 (9.2%)15,604 (10.2%).4673
  65–6994 (19.6%)38,025 (24.9%).0080
  70–74128 (26.7%)37,174 (24.3%).2263
  75–79102 (21.3%)25,731 (16.8%).0096
  80–8446 (9.6%)12,754 (8.3%).3271
  85–8915 (3.1%)2829 (1.9%).0403
  >90a2356 (1.5%)b
Sex (female), No.286 (59.7%)96,161 (62.9%).1434
Insurance, No.
  Fully insured with Humana36 (7.5%)16,579 (1.9%).0197
  Medicare447 (93.3%)135,981 (89%).0030
  Medicaid0 (0%)463 (0.3%)b
Medical comorbidity, No.
  Arrhythmia309 (64.5%)54,266 (35.5%)<.0001
  Valvular disease311 (65.3%)43,812 (28.7%)<.0001
  Chronic kidney disease154 (32.2%)40,744 (26.7%).0070
  Diabetes mellitus type 2224 (46.8%)67,700 (44.3%).2854
  Perivascular disease143 (29.9%)30,514 (2.0%)<.0001
  Stroke33 (6.9%)2239 (1.5%)<.0001
Anticoagulation, No.
  Aspirina1851 (1.2%)b
  Fondaparinuxa2052 (1.3%)b
  Heparin0 (0%)28 (0%)b
  Low-molecular weight heparin55 (11.5%)21,152 (13.8%).1356
  Warfarin114 (23.8%)31,025 (20.3%).0580
  Thrombin inhibitora179 (0.1%)b
  Factor Xa inhibitor42 (8.8%)14,448 (9.5%).6193
Length of stay, median (lower quartile, upper quartile), d3 (0, 3)3 (0, 3).2163
Authors

The authors are from the Duke University School of Medicine (CSH) and the Department of Orthopaedic Surgery (SPR, SSW, WAJ, MPB, TMS), Duke University Medical Center, Durham, North Carolina; and the Department of Orthopaedic Surgery (CP), University of California San Diego.

Ms Hong and Drs Politzer, Ryan, Wellman, and Seyler have no relevant financial relationships to disclose. Dr Jiranek has received royalties from DePuy Synthes. Dr Bolognesi is a paid consultant for TJO; has received research support from Biomet, DePuy, Zimmer, and Exactech; and receives royalties from Zimmer and TJO.

Correspondence should be addressed to: Cierra S. Hong, BA, Duke University School of Medicine, 10 Duke Medicine Circle, Durham, NC 27710 ( cierra.hong@duke.edu).

Received: November 29, 2018
Accepted: February 25, 2019
Posted Online: February 20, 2020

10.3928/01477447-20200213-06

Sign up to receive

Journal E-contents