We studied the time course of micro-embolism by ... period in 57 of 63 consecutive patients after hip or knee ..... Grady-Benson JC, Oishi CS, Hanson PB, et al.
PROLONGED ASYMPTOMATIC MICRO-EMBOLISM AFTER HIP OR KNEE ARTHROPLASTY P. ABRAHAM,
D. CARTER,
J. R. MILLOT,
G. LEFTHERIOTIS,
L. PIDHORZ,
J. L. SAUMET
From Centre Hospitalier Universitaire d’Angers, France
We studied the time course of micro-embolism by recording high intensity transient signals (HITS) on Doppler venous blood flow studies during a 7-day period in 57 of 63 consecutive patients after hip or knee replacement. No HITS were found before surgery, or in the non-operated leg after surgery. In the operated leg, the median number of HITS per minute showed an exponential decrease with time; a 50% reduction in the number of HITS took 72 minutes. Regardless of the duration and severity of HITS, we found no clinically apparent embolic event. Colour Doppler imaging at days 8 to 10 after operation showed a deep venous thrombosis in 17% of our patients with one thrombosis of the long saphenous vein. The presence or severity of HITS did not predict the venous thrombi. Although we found no pulmonary complications, we detected micro-emboli over a longer period after surgery than is usually reported. Spectral analysis of the Doppler venous signal is a repeatable and non-invasive monitoring technique in the post-surgical period. We failed to prove a correlation between deep venous thrombi and micro-emboli detected by this method. J Bone Joint Surg [Br] 1997;79-B:269-72. Received 8 February 1996; Accepted after revision 11 November 1996
Pulmonary embolism is a major complication of orthopaedic surgery. A review of the literature reported that the incidence of pulmonary embolism after hip arthroplasty was between 3% and 49%, and the number of patients who 1 had no symptoms was between 17% and 100%. After knee arthroplasty the same authors found a 5% to 21% rate of pulmonary embolism and 80% to 100% of patients were 2 asymptomatic. Recent studies with iterative lung scans 3 4 transthoracic and transoesophageal echocardiography have confirmed these findings by showing that 76% to 90% of patients with pulmonary embolism remained asymptomatic after orthopaedic surgery. Neither iterative lung scanning nor angiography are routine investigations, and transoesophageal echocardiography has potential 5 complications. The mean duration of micro-embolic events previously 3,4,5 reported ranges from 20 to 421 seconds. One study reported micro-emboli in one subject one hour after sur5 gery, but the recording was then stopped for technical reasons. Preliminary results from our laboratory have shown that most patients demonstrated micro-emboli at two 6 hours after hip or knee surgery. These micro-emboli were detected by Doppler ultrasonography of the peripheral veins which is an easy, repeatable and non-invasive technique. We have studied the prevalence and importance of micro-embolic events for a prolonged period after hip and knee replacement, and assessed the prognostic value of high intensity transient signals (HITS) for venous thrombosis. PATIENTS AND METHODS
P. Abraham, MD, PhD J. R. Millot, BSc G. Leftheriotis, MD, PhD J. L. Saumet, MD, PhD Laboratoire d’Explorations Vasculaires L. Pidhorz, MD Service de Chirurgie Orthopedique, Centre Hospitalier Universitaire d’Angers, 49033 Angers Cedex 01, France. D. Carter, BSc School of Chemical and Life Sciences, University of Greenwich, Woolwich Campus, Wellington Street, Woolwich, London, SE18 6PF, UK. Correspondence should be sent to Professor J. L. Saumet. ©1997 British Editorial Society of Bone and Joint Surgery 0301-620X/97/27132 $2.00 VOL. 79-B, NO. 2, MARCH 1997
Sixty-three consecutive patients (mean age 70.3 ± 7.9 years, mean weight 72.3 ± 12.5 kg and mean height 162.8 ± 6.3 cm) who underwent unilateral total hip or knee replacement for osteoarthritis were studied over a 3-month period. All the subjects gave informed consent to the study which had been approved by our institutional ethics committee. The 33 women and 30 men received anticoagulation prophylaxis with low molecular weight heparin using Fragmine (Pharmacia SA, Saint Quentin en Yvelines, France): 5000 U 12 hours before surgery, 2500 U following surgery and thereafter 5000 U per day. For hip surgery we used cemented PVL (Landanger, Chaumont, France) or unce269
270
P. ABRAHAM,
D. CARTER,
mented ABG (Howmedica, Lyon, France) prostheses. For knee surgery a tourniquet was applied and cemented Wallaby (Protek-Synthes, Montbeliard, France) prostheses were used. Doppler ultrasound examinations were performed during the 24 hours before surgery and one hour (H1), two hours (H2), one day (D1), two days (D2) and seven days (D7) after completion of the operation. Recording of venous blood flow in both the right and left common femoral veins was performed manually with a 4 MHz continuous wave Doppler (UV2M, Mira Electronics, Lyon, France). The best signal-to-noise ratio was sought and with the minimal possible angle between the probe axis and the vein. Signals were recorded for 4 to 5 minutes on both sides, using an audio tape recorder (Sony, DTC690, Japan). Manual blind analysis of the signal was performed off line on a spectral analyser (Spectradop, DMS, Lattes, France) and high intensity transient signals 7 defined as previously reported. Evidence of venous thrombosis in the lower limb (LLVT) was sought by trained operators using colour Doppler imaging (Acuson 128XP10 Acuson, Mountain View, USA), at between days eight and ten after surgery.
J. R. MILLOT,
ET AL
number of HITS per minute between the two groups (Table I). A significant difference in the number of HITS per minute was noted between women (n = 30) and men (n = 27) in the first and second hour after surgery, 205.0 (6.4 - 50.5) v 20.0 (0.0 - 7.5) and 122.0 (7.3 - 44.0) v 18.0 (0.0 - 5.0) respectively, but no difference was noted on days 1, 2 and 7. Results are presented as medians with the 25 and 75 centile ranges calculated. Inter-group differences of the number of HITS were studied by MannWhitney tests. Colour Doppler ultrasound imaging at one week after surgery detected LLVT in ten patients, and thrombosis of the long saphenous vein in one. Among the deep venous thrombi, no proximal (iliac or femoropopliteal) thrombi were found. Thrombosis was found in one subject in the posterior tibial vein of the non-operated leg, and in two patients in the gastrocnemius veins of the operated side. Results for the number of HITS in patients with and without LLVT are presented in Table II. No relation was found between the proportion of patients presenting HITS, and the number of HITS per minute in patients with or
RESULTS One 60-year-old woman had unstable angina pectoris, and in five patients Doppler or recorder failure prevented complete collection of data. Therefore, from the 63 patients we obtained complete data for 57. No HITS were found before surgery in any patients. Following surgery no HITS were detected at any time in the non-operated leg. Among the 57 patients micro-emboli were detected in more than half at 24 hours after surgery (Fig. 1). A few HITS were found on the seventh day in one 76-year-old man who had a deep venous thrombosis of the calf. The median number of HITS per minute in all patients showed an exponential decrease over time: 35.0 (5.0-106.0), 13.0 (0.2-60.0), 0.8 (0.0-5.0), 0.0 (0.0-1.2) at H1, H2, day 1 and day 2 respectively. Values as high as 300 to 600 HITS per minute were found at H1 and H2 in some patients who did not have clinically detectable signs of pulmonary embolism. Although the proportion of patients demonstrating HITS was greater after knee than after hip surgery at day 1 and day 2, we found no difference in the
Fig. 1 Prevalence of micro-emboli as detected by Doppler high intensity transient signals found on the common femoral vein, among a population of 57 subjects in the first week following hip or knee arthroplasty.
Table I. Percentage of patients with positive Doppler analysis and number of high intensity transient signals per minute expressed as median (25 and 75 centile) following hip or knee arthroplasty Hip surgery (n = 38)
Knee surgery (n = 19)
Patients (%)
HITS per minute
Patients (%)
HITS per minute
Hour 1 2
78.9 73.7
28.3 (1.4 to 106.0) 9.5 (0.0 to 58.5)
89.5 78.9
39.4 (6.4 to 180.0) 18.0 (0.2 to 100.0)
Day 1 2 7
42.1 31.6 2.6
0.0 (0.0 to 4.0) 0.0 (0.0 to 1.0) 0.0 (0.0 to 0.0)
73.7 57.9 0.0
3.6 (0.0 to 7.5) 0.7 (0.0 to 0.2) 0.0 (0.0 to 0.0) THE JOURNAL OF BONE AND JOINT SURGERY
PROLONGED ASYMPTOMATIC MICRO-EMBOLISM AFTER HIP OR KNEE ARTHROPLASTY
271
Table II. Percentage of patients with positive Doppler analysis and number of high intensity transient signals per minute expressed as median (25 and 75 centile) in subjects with and without lower limb venous thrombosis (LLVT) following hip or knee arthroplasty With LLVT (n = 11)
Without LLVT (n = 46)
Patients (%)
HITS per minute
Patients (%)
HITS per minute
Hour 1 2
90.9 81.8
42.0 (28.0 to 180.0) 29.0 (0.2 to 122.0)
80.4 73.9
19.0 (2.5 to 106.0) 10.0 (0.0 to 50.0)
Day 1 2 7
63.6 36.4 9.1
50.0 41.3 0.0
0.3 (0.0 to 5.0) 0.0 (0.0 to 1.4) 0.0 (0.0 to 0.0)
4.0 (0.0 to 5.5) 0.0 (0.0 to 1.2) 0.0 (0.0 to 0.0)
without LLVT. No clinically detectable complication or suspected pulmonary embolism was noted in any patient even in those with prolonged and/or frequent HITS. DISCUSSION Spectral analysis of Doppler signals has been used for years in the detection of circulating particles in animals and 8,9 humans. The sensitivity and reliability of the Doppler technique in the detection of circulating particles are 10 high. Nevertheless, there are several potential artefacts causing false-positive recordings including movements of the subject or of the probe, electronic interference and low signal-to-noise ratio. By using hand-held probes for measurements over the femoral vein a satisfactory signal-tonoise ratio was obtained in all patients, even the obese. Before surgery no patient demonstrated HITS. Soon after, embolic events of various intensity were found in more than 80%. As none of the patients in this study had recent trauma the results are consistent with recent studies 5,11 and suggest using transoesophageal echocardiography that micro-emboli are likely to be a consequence of the surgical procedure. Contrary to previous studies in which proximal recording was performed peripheral recording of venous blood flow in each leg allowed us to differentiate venous return from the operated and non-operated leg. No HITS were found in the non-operated leg following surgery confirming that the embolic source is located on the oper5 ated side as hypothesised by Christie. Although the incidence of thrombo-embolic events decreases with time the risk of developing venous thrombo-embolism may extend for several weeks after surgery. In our patients the incidence of HITS per minute from the operated leg showed an exponential decrease over time, with a mean half-time decrease lasting just over one hour. Micro-embolism seems to remain detectable for a much 3,5 longer period than previously reported. On the seventh postoperative day HITS could only be found in one 12,13 patient. This result is consistent with previous reports showing that 87% to 90% of asymptomatic embolism occurs within the first seven days after surgery and 75% 14 within the first two days. Assuming that micro-embolism detected late after surgery may be due to a local coagulative process, the decrease in HITS is consistent with either a decreased incidence of clinical complications, or a VOL. 79-B, NO. 2, MARCH 1997
15
decrease of blood coagulation markers. Some reports have stressed the risk of clinical complications as a result 5,16 However, of intraoperative or post-surgical embolism. in our study we saw no clinically detectable complications. Both the small number of subjects included, and the low sensitivity of clinical signs in the detection of pulmonary embolism may be the reason for this difference. We support the view that a direct link between asymptomatic micro-embolism and symptomatic episodes has yet to be 1 established. Venous stasis, local ischaemia, and cooling due to the use of a tourniquet have been suggested to account for the higher incidence of deep venous thrombosis in patients 11,17 after knee arthroplasty compared with hip arthroplasty. Although we found HITS more often following knee surgery there was no difference in the number of HITS after hip and knee surgery. We were not able to show a link between the presence or severity of micro-emboli and lower limb venous thrombosis. Colour Doppler imaging has a 90% sensitivity for the detection of deep venous 18,19 but there is conthrombosis in symptomatic patients, troversy about the sensitivity and specificity of ultrasound imaging in the diagnosis of asymptomatic deep venous 12,20-22 Phlebography thrombosis after orthopaedic surgery. was not performed for ethical reasons although it might have resolved the issue of low sensitivity as the explanation for absence of correlation in our study. We are also aware that phlebography may have missed the thrombi that were found in the gastrocnemius veins. In vitro, the intensity of the signals produced by microemboli has been reported to be proportional to either the 9 nature or the size of the particle. The exact nature of the particles has not been defined in the present study, but previous studies using intracardiac sampling have shown fat, platelet clumps, marrow constituents and micro-throm5 bi in the period immediately after surgery. We were not able to determine whether the micro-emboli recorded late after surgery were of the same nature as those found in the immediate post-surgical period. The Doppler detection of micro-emboli in the peripheral venous flow is safer than techniques used previously; it permits non-invasive monitoring during the period immediately after orthopaedic surgery. Because our study failed to confirm a link between asymptomatic micro-embolism and symptomatic thrombotic or embolic accidents pro-
272
P. ABRAHAM,
D. CARTER,
spective studies are needed in a larger population. We have shown that micro-emboli from the operated limb remain detectable for several hours following surgery. This is much longer than previously suggested. This work was supported in part by a grant of the European ERASHUS programme. Although none of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article, benefits have been or will be received but are directed solely to a research fund, educational institution, or other non-profit institution with which one or more of the authors is associated. REFERENCES 1. Wolf LD, Hozack WJ, Rothman RH. Pulmonary embolism in total joint arthroplasty. Clin Orthop 1993;288:219-33. 2. Haas SB, Tribus CB, Inall JN, Becker MW, Windsor RE. The significance of calf thrombi after total knee arthroplasty. J Bone Joint Surg [Br] 1992;74-B:799-802. 3. Svartling N. Detection of embolized material in the right atrium during cementation in hip arthroplasty. Acta Anaesth Scand 1988;32:203-8. 4. Christie J, Robinson CM, Singer B, Ray DC. Medullary lavage reduces embolic phenomena and cardiopulmonary changes during cemented hemiarthroplasty. J Bone Joint Surg [Br] 1995;77-B: 456-9. 5. Christie J, Robinson CM, Pell ACH, McBirnie J, Burnett R. Transcardiac echocardiography during invasive intramedullary procedures. J Bone Joint Surg [Br] 1995;77-B:450-5. 6. Abraham P, Millot JR, Pidhorz L, Saumet JL. Transcardiac echocardiography during invasive intramedullary procedures (letter). J Bone Joint Surg [Br] 1996;78-B:854-5. 7. Tegeler CH. High-intensity transient signals detected by Doppler ultrasonography: searching for answers. Cerebrovasc dis 1994;4: 379-82. 8. Austen WG, Howry DH. Ultrasound as a method to detect bubbles or particulate matter in the arterial line during cardiopulmonary bypass. J Surg Res 1965;5:283-4. 9. Bunegin L, Wahl D, Albin MS. Detection and volume estimation of embolic air in the middle cerebral artery using transcranial Doppler sonography. Stroke 1994;25:593-600.
J. R. MILLOT,
ET AL
10. Stump DA, Stein CS, Tegeler CH, Hitchings LP, et al. Validity and reliability of an ultrasound device for detecting carotid emboli. J Neuroimag 1991;21:18-22. 11. Parmet JL, Horrow JC, Singer R, Berman AT, Rosenberg H. Echogenic emboli upon tourniquet release during total knee arthroplasty: pulmonary hemodynamic changes and embolic composition. Anesth Analg 1994;79:940-5. 12. Sikorski JM, Hampson WG, Staddon GE. The natural history and aetiology of deep vein thrombosis after total hip replacement. J Bone Joint Surg [Br] 1981;63-B:171-7. 13. Westermann K, Trentz O, Pretschner P, Mellmann J. Thromboembolism after hip surgery. Int Orthop 1981;4:253-7. 14. Berqvist D, Lindblad B. Incidence of thromboembolism in medical and surgical patients. In: Bergqvist D, Comerota AJ, Nicolaides AN, Scurr JH, ed. Prevention of venous thromboembolism. Med-Orion, 1994:3-15. 15. Dahl OE, Pedersen T, Kierulf P, et al. Sequential intrapulmonary and systemic activation of coagulation and fibrinolysis during and after total hip replacement surgery. Thromb Res 1993;70:451-8. 16. Modig J, Busch C, Olerud S, Saldeen T, Waernbaum G. Arterial hypotension and hypoxaemia during total hip replacement: the importance of thromboplastic products, fat embolism and acrylic monomers. Acta Anaesth Scand 1975;19:28-43. 17. Oishi CS, Grady-Benson JC, Otis SM, Colwell CW Jr, Walker RH. The clinical course of distal deep venous thrombosis after total hip and knee arthroplasty, as determined with duplex ultrasonography. J Bone Joint Surg [Am] 1994;76-A:1658-63. 18. Mattos MA, Londrey GL, Leutz DW, et al. Color-flow duplex scanning for the surveillance and diagnosis of acute deep venous thrombosis. J Vasc Surg 1992;15:366-76. 19. Grady-Benson JC, Oishi CS, Hanson PB, et al. Postoperative surveillance for deep venous thrombosis with duplex ultrasonography after total knee arthroplasty. J Bone Joint Surg [Am] 1994;76-A: 1649-57. 20. Holden RW, Klatte EC, Park HM, et al. Efficacy of noninvasive modalities of diagnosis of thrombophlebitis. Radiology 1981; 141:63-6. 21. Davidson BL, Elliott CG, Lensing AW. Low accuracy of color Doppler ultrasound in the detection of proximal leg vein thrombosis in asymptomatic high risk patients. Ann Intern Med 1992;117:735-8. 22. Elliott CG, Suchyta M, Rose SC, et al. Duplex ultrasonography for the detection of deep vein thrombi after total hip or knee arthroplasty. Angiology 1993;44:26-33.
THE JOURNAL OF BONE AND JOINT SURGERY
