Patients undergoing surgical procedures in general and those receiving total joint arthroplasty in particular are at risk of having pulmonary embolus (PE) and deep venous thrombosis develop.5,15,20,24,27,29,32,48,49 A high incidence of thromboembolism, including fatal PE, has been reported in patients undergoing total joint arthroplasty who did not receive prophylaxis.2,45 Because of the latter, administration of mechanical and/or chemical prophylaxis after joint arthroplasty has become a common practice. 1,7-9,11-13,18,19,22,30,33,38,39 Although anticoagulation has been effective in reducing the incidence of thromboembolism, deaths secondary to PE after joint arthroplasty still occur.21,26,52 Currently, nearly one in 1000 patients undergoing joint replacement is expected to die of PE.21,52 Although our perioperative mortality rate (within 30 days) from PE and nonPE has remained stable at approximately 0.1 to 0.2 during the last few years,43 we have observed an increase in the incidence of diagnosed nonfatal PE. This is despite the fact our anticoagulation protocol and surgical practice remained unaltered during the study period. This observation provided the impetus for our study attempting to elucidate the potential reasons for the increase in the incidence of nonfatal PE.
We hypothesized modern sophisticated imaging with better capability to detect small emboli may be the major reason for the suspected increase in the incidence of non-fatal PE. Second, we asked if other factors such as patients' demographic characteristics and the type of surgery may influence the incidence of PE. Last but not least, we looked if the outcome of these patients (death) changed accordingly to the increased incidence of PE.
MATERIALS AND METHODS
We performed a retrospective comparative study to examine the factors that may be responsible for the increase in the incidence of PE at our center during recent years. Several factors were analyzed, including the influence of patients' gender, age, body mass index, type of surgery, joint, unilateral versus bilateral procedures, and the imaging modality used (VQ scan or computed tomography [CT]) for diagnosis. Using our prospective database, we identified 14,890 joint arthroplasties performed between January 2000 and December 2005 in 13,133 patients. Of these, 6950 cases (5695 patients) were knee and 7940 cases (7438 patients) were hip arthroplasties. We then identified 145 patients (1.10%) with image-confirmed diagnoses of PE after joint arthroplasty performed between 2000 and 2005.
Of these, 44 (30.3%) were men and 101 (69.7%) were women with a minimum age of 36.2 years (mean, 69.6 years; range, 36.2-90.0 years). The minimum weight and height of the patients was 41.7 kg (mean, 90.1 kg; range, 41.7-167.6 kg) and 63.5 cm (mean, 166.6 cm; range, 63.5-221.0 cm), respectively, corresponding to a minimum body mass index of 15.8 kg/m2 (mean, 31.98 kg/m2; range, 15.8-54.0 kg/m2).
We perform hip arthroplasty with the patient in the supine position. An anterolateral approach was used in all hip cases. Knee arthroplasty is performed using a tourniquet with medial parapatellar arthrotomy.
The prophylactic anticoagulation regimen for patients undergoing joint arthroplasty was the same throughout the study period. This regimen in general consisted of a 5- to 10-mg warfarin loading dose administered on the night of the procedure followed by daily dosing aiming for an international normalized ratio between 1.5 and 1.7. All patients received anticoagulation for 6 weeks. We used additional strategies to prevent thromboembolic events. Unless contraindicated, patients received 1000 units of intravenous heparin at the time of hip dislocation or before inflation of the tourniquet during knee arthroplasties. The majority (greater than 90%) of elective arthroplasties were performed under hypotensive regional anesthesia. In addition, we made efforts to perform the surgery in an expeditious manner. Patients had early and aggressive postoperative mobilization by the nursing and physical therapy staff. We used continuous passive motion devices after all knee arthroplasties.
During the study period, patients with unexpected lower extremity swelling, calf tenderness, and calf pain had Doppler ultrasound to detect deep venous thrombosis. Patients who had signs and symptoms suspicious for PE (chest pain, shortness of breath, tachycardia, tachypnea, diminished pulse oximetry reading) initially were evaluated with an electrocardiogram, chest radiograph, cardiac enzymes, arterial blood gas, blood chemistry, and electrolytes. An internist was consulted at this point.
The imaging modality of choice for investigation of PE changed during the study period. Until May 2001, the imaging modality for investigation of PE was VQ scan. We obtained pulmonary angiography for cases with low or moderate probability of PE based on the VQ scan for which a high clinical suspicion was maintained. A normal VQ scan was considered to essentially rule out a PE.23,28 VQ scans still are used today in patients with renal failure or allergy to intravenous dye.
Beginning in May 2001, the VQ scan was almost completely replaced by spiral CT to confirm PE. Spiral CT was fast, easily accessible, and offered excellent visualization of the main and lobar branches of the pulmonary vasculature. Because of its relatively low sensitivity (21-57%)36 to detect segmental and subsegmental emboli, the imaging technology continued to advance and spiral CT was upgraded to multidetector CT (MDCT) in November 2003.
With MDCT, the xray beam is registered by multiple-row (4, 16, 40, or 64) detector arrays rather than a single row like the original spiral CT. The increased number of detector arrays allows faster scanning of a specified area. Multiple thin slices and increased speed improve spatial and temporal resolution. Also, a relatively higher dose of radiation can be used, because it is used over a shorter period of time, which decreases image noise and further increases resolution. These improvements result in dramatically enhanced resolution, especially of the smaller and more peripheral vessels.41
Patients diagnosed with PE received anticoagulation treatment for 6 months. Intravenous heparin was given initially untiloral anticoagulation had reached therapeutic levels (international normalized ratio goal, 2.5-3). We used an inferior vena cava filter in 134 of the 145 (92.4%) patients. This modality frequently was chosen if the PE occurred during the first 2 postoperative weeks. On occasion, we individualized the anticoagulation regimen.
All patients were evaluated at 6 weeks, 6 months, and 2 years postoperatively. Patients were encouraged to return every 2 years thereafter. If unable to report for a physical examination, the patients were contacted by telephone. When unable to contact the patient, a private investigator's site (www.intelius.com) was used to access patients' contact details. The Social Security Death Index (http://ssdi.rootsweb.com) also was checked frequently to identify patients who had died who were not in our prospective joint registry. We contacted or accounted for all patients and their living status.
We (JM, WM, LP) reviewed in detail all images (VQ scan, spiral CT, and MDCT) performed for the investigation of patients with suspected PE. The entire lung field on the images was examined to determine the presence or absence of emboli. When present, the size and the exact anatomic location of the emboli were determined.
We analyzed individual risk factors for PE, including age, body mass index, type of procedure, joint affected, unilateral versus bilateral, year of surgery, and the imaging modality used for diagnosis. Fisher's exact test was used to evaluate categorical variables, whereas two-sample t test was used for evaluation of continuous variables. A 95% confidence interval was used for significance. All analyses were performed using SPSS (version 13; SPSS, Chicago, IL).
The overall incidence of diagnosed PE during our study was 1.10% (145 of 13,133) (Table 1). The incidence of diagnosed PE during the time when VQ scan was being used was 0.21% (six of 2859) compared with 0.98% (50 of 5095) during the time when single-slice spiral CT was used and 1.72% (89 of 5179) during the latter years of the study when MDCT was used to investigate PE. Although the number of patients undergoing joint arthroplasty per year increased (1.23 times) from 1946 in 2000 to 2400 in 2005 (Fig 1), the incidence of diagnosed PE increased ninefold during the same period, confirming the increase in the incidence of diagnosed PE was not purely the result of an increase in the number of arthroplasties.
Before May 15, 2001, when VQ scan was the modality of choice, emboli were evenly diagnosed in the lobar (three of six) and segmental (three of six) pulmonary arteries. When spiral CT was used (May 15, 2001, through October 31, 2003), 63.6% of the emboli were diagnosed in the main (11 of 44) and lobar (17 of 44) arteries and 36.4% in the segmental (14 of 44) and subsegmental (two of 44) pulmonary arteries. After October 31, 2003, there was an increase (p < 0.0001) in the number (64) and percentage (75.8%) of PEs being diagnosed in the segmental (54 of 85) or subsegmental (10 of 85) vasculature by MDCT, attesting to the ability of the latter imaging modality to detect smaller emboli (Fig 2).
There were wide variations in the incidence based on the year of surgery and the type of arthroplasty performed. There was an increase (p < 0.0001) in the incidence of diagnosed PE with time from 0.21% (four of 1946) in 2000 to 1.50% (36 of 2400) in 2005 (Fig 1). However, the number of CT studies increased almost 400% from 21 in 2001 to 81 in 2005, and the percent of positive tests increased during the same time from 38.1% to 44.4% (Fig 3).
Patients who had PE develop generally were older (p < 0.0001) (mean, 69.6 years; range, 36.2-90.0 years) compared with the overall population (mean, 64.1 years; range, 12-103 years). The body mass index of patients who had PE was higher (p = 0.001) (mean, 31.98 kg/m2; range, 15.8-54.0 kg/m2) compared with the body mass index of the overall population (mean, 29.95 kg/m2; range, 16.0-68.9 kg/m2).
The incidence of diagnosed PE also varied based on the type of surgical procedure the patient underwent. The incidence of PE after primary knee arthroplasty (1.90%) was higher than the incidence of PE after primary hip arthroplasty (0.58%) (p < 0.001) (Fig 4A). The incidence of PE after all knee arthroplasties combined at 1.81% was higher (p < 0.0001) than all hip arthroplasties combined at 0.56% (Fig 4A). We observed a similar incidence of diagnosed PE after primary (1.18%) and revision (0.73%) procedures. The incidence of diagnosed PE after bilateral procedures (2.11%) performed under the same anesthesia was also higher (p < 0.0001) than unilateral procedures (0.95%) for hip and knee arthroplasties (Fig 4B).
We identified three deaths related to PE accounting for an overall incidence of 0.023% for fatal PE. All deaths occurred in the hospital. Pulmonary embolism was diagnosed by CT scan in two of the patients and by VQ scan in the third. Autopsy was not performed on any of these patients because of family wishes. Despite an increase in the incidence of PE with time, there was no change in the incidence of 90-day mortality during our study.
During the last few years, despite making no alteration in surgical, anesthesia, or perioperative protocols, the incidence of diagnosed PE has increased markedly at our institution. The number of diagnosed PE has increased ninefold during a 5-year period raising serious concerns. The patient safety committee in our hospital tracking this trend called for reexamination of the current anticoagulation regimen and possible implementation of changes in line with recent published recommendations.13 Before departing from our established anticoagulation protocols with more than 20 years of proven safety and efficacy, we decided to investigate this matter further.
This study has several limitations. The study design, as a retrospective analysis of prospectively collected data, may reduce but not avoid possible recall and selection bias. Data collection and analysis of the study population did not include all the factors that may play a role in our study question. We did not include predisposing factors for PE such as previous deep vein thrombosis or PE, comorbidities, level of anticoagulation, and blood transfusions. However, we believe this would not affect our conclusions because there were no changes in our patient population or in patient care during the study period. Differentiating fat, cement, and thrombotic embolism was not possible. Therefore, we assumed patients with a diagnosis of PE had a thrombotic event, although this may overestimate the number. Nonetheless, although we believe we have identified the majority of embolic events, it is impossible to estimate the incidence of nonfatal silent or missed diagnosed PE.
We found the number of imaging investigations ordered for diagnosis of potential PE increased during the 5-year study period. The reason for the latter appeared multifactorial. With the improvements in nursing care and administration of long-acting intrathecal opioids, recording of pulse oximetry after total joint arthroplasty has become a common practice. Therefore, any drop in oxygen saturation triggered medical consultations and possible investigations for PE. The other reason for the increase in the number of investigations performed related to logistics of ordering and performing such tests. Whereas performing VQ scans required preparation, precluding its administration during the weekend or off hours at our institution, CT could be performed with relative ease and at any time.
We also observed the increase in diagnosed PE corresponded with the type of investigative imaging. The introduction of a more sophisticated imaging modality had resulted in a higher incidence of detected PE. CT scanners have been proposed as a first-line test for the diagnosis of PE.44 They have become more sensitive with the evolution from helical (spiral) CT to the more sophisticated MDCT in detecting small emboli.14,31,34,35,41 Other studies have confirmed more segmental and subsegmental emboli are detected by MDCT than spiral CT and VQ scan.4,35,41 However, regardless of the size and location of the PE, patients had prolonged anticoagulation treatment and/or inferior vena cava filter insertion.
Our data are consistent with the notion that the apparent increase in the incidence of diagnosed PE relates to the increase in the number of investigations and the improvement in the sensitivity of imaging modalities being used to diagnose PE. Thus, the increase in the incidence of diagnosed PE is a relative phenomenon because a large number of patients being diagnosed with PE currently would have not been diagnosed in the past because of not having an imaging investigation or having imaging with a less sensitive modality. We are confident regarding the latter statement because we have, despite implementing a much more stringent followup in recent years, not noted an increase in the incidence of fatal PE.
It is established that imaging tests such as MDCT with a relatively high sensitivity (90%, 96%, and 100%) and specificity (86%, 89%, and 94)6,34,50 are more likely to detect an embolus than an older imaging modality such as VQ scan. Multidetector CT currently is the preferred imaging modality for investigation of suspected PE.3,10,37,43
Multidetector CT has an exponential improvement in volume coverage speed with greater diagnostic image quality when compared with conventional spiral CT.17 It is capable of detecting and localizing small radiodense objects, which easily can be missed on axial CT.46 The fact is some of these objects, labeled as emboli, may in fact not be thrombotic in nature.16 Although the diagnosis of fat embolism by CT has been described,25 we were not able to elucidate the exact nature of the emboli seen in the segmental and subsegmental pulmonary vasculature on the CT images because no contrast medium was administered to all the patients.
We should be aware of the challenges sophisticated imaging modality in general, and modern imaging for PE in particular, will introduce to the surgical community. The introduction of MDCT started a new era in the diagnosis of PE. Computed tomography pulmonary angiography with high sensitivity and specificity is becoming the preferred imaging modality for evaluation of patients with suspected PE.40,42,47,51 Numerous institutions may witness an increase in the incidence of diagnosed PE that may be attributable to the marvel of modern imaging.
1. Anderson FA Jr, Hirsh J, White K, Fitzgerald RH Jr; Hip and Knee Registry Investigators. Temporal trends in prevention of venous thromboembolism following primary total hip or knee arthroplasty 1996-2001: findings from the Hip and Knee Registry. Chest
. 2003;124 (suppl):349S-356S.
2. Ansari S, Warwick D, Ackroyd CE, Newman JH. Incidence of fatal pulmonary embolism after 1,390 knee arthroplasties without routine prophylactic anticoagulation, except in high-risk cases. J Arthroplasty
3. Baile EM, King GG, Muller NL, D'Yachkova Y, Coche EE, Paré PD, Mayo JR. Spiral computed tomography is comparable to angiography for the diagnosis of pulmonary embolism. Am J Respir Crit Care Med
4. Blachere H, Latrabe V, Montaudon M, Valli N, Couffinhal T, Raherisson C, Leccia F, Laurent F. Pulmonary embolism revealed on helical CT angiography: comparison with ventilation-perfusion radionuclide lung scanning. AJR Am J Roentgenol
5. Borghi B, Casati A; Rizzoli Study Group on Orthopaedic Anaesthesia. Thromboembolic complications after total hip replacement. Int Orthop
6. Coche E, Verschuren F, Keyeux A, Goffette P, Goncette L, Hainaut P, Hammer F, Lavenne E, Zech F, Meert P, Reynaert MS. Diagnosis of acute pulmonary embolism in outpatients: comparison of thin-collimation multi-detector row spiral CT and planar ventilation-perfusion scintigraphy. Radiology
7. Colwell CW Jr, Collis DK, Paulson R, McCutchen JW, Bigler GT, Lutz S, Hardwick ME. Comparison of enoxaparin and warfarin for the prevention of venous thromboembolic disease after total hip arthroplasty: evaluation during hospitalization and three months after discharge. J Bone Joint Surg Am
8. Douketis JD, Eikelboom JW, Quinlan DJ, Willan AR, Crowther MA. Short-duration prophylaxis against venous thromboembolism after total hip or knee replacement: a meta-analysis of prospective studies investigating symptomatic outcomes. Arch Intern Med
9. Eikelboom JW, Quinlan DJ, Douketis JD. Extended-duration prophylaxis against venous thromboembolism after total hip or knee replacement: a meta-analysis of the randomised trials. Lancet
10. Filipek MS, Gosselin MV. Multidetector pulmonary CT angiography: advances in the evaluation of pulmonary arterial diseases. Semin Ultrasound CT MR
11. Fitzgerald RH Jr, Spiro TE, Trowbridge AA, Gardiner GA Jr, Whit-sett TL, O'Connell MB, Ohar JA, Young TR; Enoxaparin Clinical Trial Group. Prevention of venous thromboembolic disease following primary total knee arthroplasty: a randomized, multicenter, open-label, parallel-group comparison of enoxaparin and warfarin. J Bone Joint Surg Am
12. Freedman KB, Brookenthal KR, Fitzgerald RH Jr, Williams S, Lonner JH. A meta-analysis of thromboembolic prophylaxis following elective total hip arthroplasty. J Bone Joint Surg Am
13. Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, Colwell CW, Ray JG. Prevention of venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest
. 2004;126(3 suppl):338S-400S.
14. Ghaye B, Szapiro D, Mastora I, Delannoy V, Duhamel A, Remy J, Remy-Jardin M. Peripheral pulmonary arteries: how far in the lung does multi-detector row spiral CT allow analysis? Radiology
15. Giuntini C, Di Ricco G, Marini C, Melillo E, Palla A. Pulmonary embolism: epidemiology. Chest
. 1995;107(1 suppl):3S-9S.
16. Han D, Lee KS, Franquet T, Müller NL, Kim TS, Kim H, Kwon OJ, Byun HS. Thrombotic and nonthrombotic pulmonary arterial embolism: spectrum of imaging findings. Radiographics
17. Hu H, He HD, Foley WD, Fox SH. Four multidetector-row helical CT: image quality and volume coverage speed. Radiology
18. Imperiale TF, Speroff T. A meta-analysis of methods to prevent venous thromboembolism following total hip replacement. JAMA
19. Kearon C. Duration of venous thromboembolism prophylaxis after surgery. Chest
. 2003;124(6 suppl):386S-392S.
20. Kim YH, Oh SH, Kim JS. Incidence and natural history of deep-vein thrombosis after total hip arthroplasty: a prospective and randomised clinical study. J Bone Joint Surg Br
21. Lie SA, Engesaeter LB, Havelin LI, Furnes O, Vollset SE. Early postoperative mortality after 67,548 total hip replacements: causes of death and thromboprophylaxis in 68 hospitals in Norway from 1987 to 1999. Acta Orthop Scand
22. Lieberman JR, Hsu WK. Prevention of venous thromboembolic disease after total hip and knee arthroplasty. J Bone Joint Surg Am
23. Macdonald WB, Patrikeos AP, Thompson RI, Adler BD, van der Schaaf AA. Diagnosis of pulmonary embolism: ventilation perfusion scintigraphy versus helical computed tomography pulmonary angiography. Australas Radiol
24. Mahomed NN, Barrett JA, Katz JN, Phillips CB, Losina E, Lew RA, Guadagnoli E, Harris WH, Poss R, Baron JA. Rates and outcomes of primary and revision total hip replacement in the United States medicare population. J Bone Joint Surg Am
25. Malagari K, Economopoulos N, Stoupis C, Daniil Z, Papiris S, Müller NL, Kelekis D. High-resolution CT findings in mild pulmonary fat embolism. Chest
26. Mantilla CB, Horlocker TT, Schroeder DR, Berry DJ, Brown DL. Frequency of myocardial infarction, pulmonary embolism, deep venous thrombosis, and death following primary hip or knee arthroplasty. Anesthesiology
27. Mantilla CB, Horlocker TT, Schroeder DR, Berry DJ, Brown DL. Risk factors for clinically relevant pulmonary embolism and deep venous thrombosis in patients undergoing primary hip or knee arthroplasty. Anesthesiology
28. Marini C, Palla A, Giuntini C. Pulmonary embolism: lung scan and computed tomography. Ital Heart J
29. Mont MA, Jones LC, Rajadhyaksha AD, Shuler MS, Hungerford DS, Sieve-Smith L, Wang P, Cordista AG, Glueck CJ. Risk factors for pulmonary emboli after total hip or knee arthroplasty. Clin Orthop Relat Res
30. Paiement GD, Wessinger SJ, Hughes R, Harris WH. Routine use of adjusted low-dose warfarin to prevent venous thromboembolism after total hip replacement. J Bone Joint Surg Am
31. Patel S, Kazerooni EA, Cascade PN. Pulmonary embolism: optimization of small pulmonary artery visualization at multi-detector row CT. Radiology
32. Phillips CB, Barrett JA, Losina E, Mahomed NN, Lingard EA, Guadagnoli E, Baron JA, Harris WH, Poss R, Katz JN. Incidence rates of dislocation, pulmonary embolism, and deep infection during the first six months after elective total hip replacement. J Bone Joint Surg Am
33. Pitto RP, Hamer H, Heiss-Dunlop W, Kuehle J. Mechanical prophylaxis of deep-vein thrombosis after total hip replacement: a randomised clinical trial. J Bone Joint Surg Br
34. Qanadli SD, Hajjam ME, Mesurolle B, Barré O, Bruckert F, Joseph T, Mignon F, Vieillard-Baron A, Dubourg O, Lacombe P. Pulmonary embolism detection: prospective evaluation of dual-section helical CT versus selective pulmonary arteriography in 157 patients. Radiology
35. Raptopoulos V, Boiselle PM. Multi-detector row spiral CT pulmonary angiography: comparison with single-detector row spiral CT. Radiology
36. Rathbun SW, Raskob GE, Whitsett TL. Sensitivity and specificity of helical computed tomography in the diagnosis of pulmonary embolism: a systematic review. Ann Intern Med
37. Russo V, Piva T, Lovato L, Fattori R, Gavelli G. Multidetector CT: a new gold standard in the diagnosis of pulmonary embolism? State of the art and diagnostic algorithms. Radiol Med (Torino)
38. Salvati EA, Pellegrini VD Jr, Sharrock NE, Lotke PA, Murray DW, Potter H, Westrich GH. Recent advances in venous thromboembolic prophylaxis during and after total hip replacement. J Bone Joint Surg Am
39. Sarmiento A, Goswami AD. Thromboembolic prophylaxis with use of aspirin, exercise, and graded elastic stockings or intermittent compression devices in patients managed with total hip arthroplasty. J Bone Joint Surg Am
40. Schoepf UJ, Goldhaber SZ, Costello P. Spiral computed tomography for acute pulmonary embolism. Circulation
41. Schoepf UJ, Holzknecht N, Helmberger TK, Crispin A, Hong C, Becker CR, Reiser MF. Subsegmental pulmonary emboli: improved detection with thin-collimation multi-detector row spiral CT. Radiology
42. Stein PD, Kayali F, Olson RE. Trends in the use of diagnostic imaging in patients hospitalized with acute pulmonary embolism. Am J Cardiol
43. Tarity TD, Herz AL, Parvizi J, Rothman RH. Ninety-day mortality after hip arthroplasty: a comparison between unilateral and simultaneous bilateral procedures. J Arthroplasty
. 2006;21(suppl 2):60-64.
44. Trowbridge RL, Araoz PA, Gotway MB, Bailey RA, Auerbach AD. The effect of helical computed tomography on diagnostic and treatment strategies in patients with suspected pulmonary embolism. Am J Med
45. Warwick D, Williams MH, Bannister GC. Death and thromboembolic disease after total hip replacement: a series of 1162 cases with no routine chemical prophylaxis. J Bone Joint Surg Br
46. Weiss CR, Fishman EK. Multidetector 3D CT of pulmonary embolism of a peripheral intravenous line. Emerg Radiol
47. Weiss CR, Scatarige JC, Diette GB, Haponik EF, Merriman B, Fishman EK. CT pulmonary angiography is the first-line imaging test for acute pulmonary embolism: a survey of US clinicians. Acad Radiol
48. Westrich GH, Haas SB, Mosca P, Peterson M. Meta-analysis of thromboembolic prophylaxis after total knee arthroplasty. J Bone Joint Surg Br
49. White RH, Romano PS, Zhou H, Rodrigo J, Bargar W. Incidence and time course of thromboembolic outcomes following total hip or knee arthroplasty. Arch Intern Med
50. Winer-Muram HT, Rydberg J, Johnson MS, Tarver RD, Williams MD, Shah H, Namyslowski J, Conces D, Jennings SG, Ying J, Trerotola SO, Kopecky KK. Suspected acute pulmonary embolism: evaluation with multi-detector row CT versus digital subtraction pulmonary arteriography. Radiology
51. Wittram C, Meehan MJ, Halpern EF, Shepard JA, McLoud TC, Thrall JH. Trends in thoracic radiology over a decade at a large academic medical center. J Thorac Imaging
52. Wroblewski BM, Siney PD, Fleming PA. Fatal pulmonary embolism and mortality after revision of failed total hip arthroplasties. J Arthroplasty