Fat embolism to the pulmonary circulation is known to occur during total hip arthroplasty, especially during insertion of a cemented femoral component. Fat and air bubbles may enter the systemic circulation via a patent foramen ovale or through pulmonary circulation.


To determine whether microemboli to the brain were occurring during total hip arthroplasty, 23 patients underwent transcranial Doppler assessment of emboli to the middle cerebral artery during total hip arthroplasty. Surgery was performed with the patient in the lateral decubitus position so that the probe recorded from the nondependent side.


Successful recordings were made in 20 patients, in 8 of 20 patients there were embolic signals, which ranging from 1 to 200. In all eight patients, signals were recorded during impaction of a cemented component or after relocation of the hip. Only one patient showed evidence of emboli with impaction of the acetabulum component. In two patients there were 150 and 200 embolic signals: in both mild respiratory symptoms developed. One patient became overtly agitated during a flurry of emboli.


Cerebral microemboli can occur during total hip arthroplasty. Whether this contributes to changes in postoperative cognitive function is unknown.

FAT embolism is known to occur after long-bone fracture and total hip arthroplasty (THA). 1,2Showers of emboli (thought to be fat and thromboemboli both) can be seen by imaging of the right heart with use of transesophageal echocardiography during THA. 2–4It is usually assumed that these emboli are trapped in the lung and do not reach the systemic circulation. On rare occasions, stroke may occur after THA, and this has been ascribed to paradoxical embolism via  a patent foramen ovale (PFO). 3,5,6Transpulmonary passage of small emboli, both fat and air, has recently been shown in experimental models. 7This introduced the possibility that the microemboli entering the lung during THA could traverse the lung and undergo embolization into the systemic circulation. The following case describes this possible phenomenon.

A 43-yr-old woman was admitted for revision of a right THA. Medical history was remarkable for severe polyarticular rheumatoid arthritis, hypertension, gastroesophageal reflux, and peptic ulcer disease. Previous surgery included bilateral total knee arthroplasties, bilateral THAs, bilateral total shoulder arthroplasties, bilateral ankle fusions, cholecystectomy, and total abdominal hysterectomy. Medications at the time of surgery were methylprednisolone, azathioprine, furosemide, methotrexate, and diltiazem.

After routine monitoring devices were applied, a radial artery catheter was inserted for continuous blood pressure determination, and a pulmonary artery catheter was inserted for cardiovascular and volume monitoring. A 2-MHz transcranial Doppler (TCD) probe was applied over the right temporal window for continuous flow velocity and embolic signal determinations, as monitored by a dedicated technician.

During pressurization and cementation of the femoral canal with use of the femoral prosthesis, mean pulmonary artery pressure increased from 13 mmHg to 20 mmHg. Coincidentally, a shower of embolic signal was noted. Clinically, the patient became agitated and required additional sedation.

Postoperative course was remarkable for development of symptoms of chest heaviness and a decrease in partial pressure of oxygen (PO2) to 80 mmHg with administration of 4 l/min oxygen via  nasal cannula. On the fifth postoperative day, a ventilation–perfusion scan was read as low probability. The patient was discharged on postoperative day 15 with routine follow-up.

Transcranial Doppler has been used extensively to assess microemboli to the brain during open heart surgery 8and carotid angioplasty 9and in medical patients at risk for embolic stroke. 10In many patients, these microemboli produce no apparent clinical sequelae, 11but, in cardiac surgery, patients with high numbers of emboli are more likely to experience stroke perioperatively. 8,12In addition, it has been shown that emboli are most likely to occur during certain phases of the surgery, such as after removal of aortic clamps. 13 

In this preliminary study, we measured TCD signals from the middle cerebral artery in patients during THA and were able to detect the presence of cerebral emboli, primarily during insertion of cemented femoral components. This study introduces the possibility that systemic emboli are not uncommon during THA and may account for some of the cognitive impairment after total joint arthroplasty. 14,15 

After Institutional Review Board approval, 23 patients undergoing THA were prospectively studied. As is routine in this institution, after application of monitoring devices (i.e. , electrocardiography, oxygen saturation as measured by pulse oximetry [SpO2], noninvasive blood pressure) and supplemental oxygen via  nasal cannula, the patients received intravenous sedation with use of midazolam and fentanyl. Radial artery catheters were inserted in all patients. However, central venous pressure catheters and pulmonary artery catheters were inserted only if clinically indicated. The patients were placed in the lateral decubitus position, operative side up, and received a lumbar epidural (L2–L3, L3–L4) with use of either 0.75% bupivacaine or 2% lidocaine to obtain a sensory level of T2–T4. As described in previous studies, 16mean arterial pressure was maintained at 50–60 mmHg, with use of infusions of 1–5 μg/min epinephrine. Intravenous lactated Ringer’s solution was administered as needed. No patient received blood or colloid during the observation period. Antibiotics were administered before surgery, and intravenous injections were not performed during critical observation periods.

The middle cerebral artery “window” was identified preoperatively and, again, immediately before surgery. With the patient in the lateral decubitus position, the operative side uppermost, the middle cerebral artery was again determined and the probe secured with use of a headband. The epidural was then inserted and the patient remained in this position throughout surgery. A TCD model TC2-64A (Nicolet, Madison, WI) with a 2-MHz probe was used in 16 patients and a TCD (Medsonics, Mountain View, CA) in the other 4 patients. Recordings were made at a depth of 40–50 mm. A dedicated technician recorded the signal continuously during surgery. Embolic signals (ESs) were recorded manually by the technician and verified by inspection of recorded tracings. ESs  were defined as high-amplitude, unidirectional, transient signals lasting less than 0.1 s and associated with a characteristic chirping sound. 17ESs were copied to and stored on floppy disks and were subsequently reviewed and manually counted by a single examiner. The number of aggregate ESs was concurrently recorded by an automated counter. This number was only used where the series of embolization made manual counting impossible. The occurrence of signals was recorded in relation to surgical events: (1) reaming the acetabulum, (2) impaction of a porous-coated acetabular component, (3) reaming the femoral canal, (4) impaction of a cemented femoral component, and (5) final relocation of the joint after implantation of the femoral component. Recordings were discontinued at the end of surgery.

Patients did not undergo echocardiography to detect the presence of a PFO, nor were cognitive function studies performed postoperatively.

Three patients were excluded because a satisfactory assessment of the middle cerebral artery was not obtained; therefore, recordings were obtained for 20 patients. The mean age was 62 yr (range, 37–80 yr). Two patients were classified as American Society of Anesthesiologists physical status I, 15 as physical status II, and 3 as physical status III. There were 13 women and 7 men patients. ESs were detected in 8 of 20 patients. A recording of an isolated ES is shown in figure 1. The number of embolic events varied from 1 to 200 during surgery. Four of 11 patients undergoing unilateral primary THA, two of five undergoing revision of a femoral component, and both patients undergoing one-stage bilateral THA experienced embolic events. ESs were not exhibited in either patient undergoing revision of an acetabular cup alone. The mean number of ESs varied according to the procedure, as shown in table 1. ESs were not noted during reaming of the acetabulum and were noted only in one patient during impaction of the cup. ESs were noted during impaction of the cemented component in five patients and during relocation of the hip joint in four patients.

Fig. 1. A transcranial Doppler sonogram of the middle cerebral artery during total hip arthroplasty. Embolic signals are depicted in red.

Fig. 1. A transcranial Doppler sonogram of the middle cerebral artery during total hip arthroplasty. Embolic signals are depicted in red.

Table 1. Patient Demography and Number of Embolic Signals

CAD = coronary artery disease; HTN = hypertension; MVP = mitral valve prolapse; NIDDM = non-insulin dependent diabetes mellitus; RA = rheumatoid arthritis.

Table 1. Patient Demography and Number of Embolic Signals
Table 1. Patient Demography and Number of Embolic Signals

More than 150 ESs were seen in two patients. In one patient in whom a cemented femoral component was reinserted, mean pulmonary artery pressure increased from 13 to 20 mmHg as the hip joint was reduced. Immediately thereafter, a shower of emboli was detected with use of TCD sonography. The patient became restless at this time and required additional sedation, but did not exhibit any change in oxygen saturation. After surgery, the patient became slightly hypoxemic, breathing nasal oxygen at 4 l/min (p  H 7.46, partial pressure of arterial carbon dioxide [PaCO2] 34 mmHg, PO280 mmHg), had symptom of pleuritic chest pain, but the ventilation–perfusion scan showed negative findings. The other patient with more extensive embolic ESs underwent primary THA. Intraoperatively, an embolic shower of 200 ESs was noted. After surgery, mild chest pain and heaviness developed in the patient, but the ventilation–perfusion scan was normal. None of the 20 patients exhibited signs of confusion or stroke after surgery. Transpulmonary passage of these emboli may be consistent with the constellation of symptoms one can see with fat embolism syndrome.

With use of TCD sonography, embolic signals in the middle cerebral artery were detected during THA in 8 of 20 patients studied. The majority of signals occurred either during insertion of the cemented femoral component or immediately after relocation of the hip joint. It is these particular surgical maneuvers that are also associated with intraoperative pulmonary emboli, as detected by echocardiography, 2,4or with markers of thrombosis. 18For this reason, we believe these signals represent fat or thrombi in the middle cerebral artery, which have bypassed the lung either through a PFO 3or by transpulmonary passage. 7 

Intraoperative pulmonary embolization during total hip arthroplasty has been recognized for more than 20 yr but has recently been better characterized with use of intraoperative echocardiography. The emboli may be fine (particularly during reaming the femur) or large after implantation of cemented femoral components. 3PFO may be present in up to 20% of patients and may allow emboli to pass into the left atrium when right-sided pressures exceed the left. 19With pulmonary embolization, right-sided pressures increase, 18and it is curious why stroke does not develop in more patients after total joint arthroplasty. Presumably, the passage of the emboli via  a PFO is uncommon, or else these emboli are better tolerated by the cerebral circulation than one might imagine.

Recently, transpulmonary passage of fat emboli has been observed in dogs. 7This raises the possibility that the emboli we noted could have traversed the lung. Factors facilitating transpulmonary passage are unknown, but acute pulmonary hypertension caused by the emboli may facilitate passage of tiny emboli through the lung into the systemic circulation.

The clinical significance of these findings is unclear because neurologic problems did not develop after surgery in any of the patients with evidence of emboli. This is not entirely surprising because patients undergoing cardiac surgery are unlikely to experience cognitive deterioration unless 100–200 ESs occur during surgery. 8,12It is not clear whether these emboli noted in this study are well-tolerated by the brain (as may occur with air rather than plaque 11) or whether the number of emboli is too small to result in cognitive deterioration. It is, however, possible that these emboli may lead to postoperative confusion in susceptible patients (e.g. , elderly patients) and may also account for a rather high rate of persistent cognitive deficit noted in patients after total knee arthroplasty. 15,20Further studies are needed to clarify this.

Patterson BM, Healy JH, Cornell CN, Sharrock NE: Cardiac arrest during hip arthroplasty with a cemented long-stem component. J Bone Joint Surg [Am] 1991; 73-A:271–7
Woo R, Minster GJ, Fitzgerald RH Jr, Mason LD, Lucas DR, Smith FE: Pulmonary fat embolism in revision hip arthroplasty. Clin Orthop 1995; 319:41–53
Christie J, Burnett R, Potts HR, Pell ACH: Echocardiography of transatrial embolism during cemented and uncemented hemiarthroplasty of the hip. J Bone Joint Surg Br 1994; 76-B:409–12
Ereth MH, Weber JG, Abel MD, Lennon RL, Lewallen DG, Ilstrup DM, Rehder K: Cemented versus noncemented total hip arthroplasty—Embolism, hemodynamics, and intrapulmonary shunting. Mayo Clin Proc 1992; 67:1066–74
Pell ACH, Christie J, Keating JF, Sutherland GR: The detection of fat embolism by transesophageal echocardiography during reamed intramedullary nailing. J Bone Joint Surg Br 1993; 75-B:921–5
Weiss SJ, Cheung AT, Stecker MM, Garino JP, Hughes JE, Murphy FL Jr: Fatal paradoxical cerebral embolization during bilateral knee arthroplasty. A nesthesiology 1996; 84:721–3
Byrick RJ, Mullen JB, Mazer CD, Guest CB: Transpulmonary systemic fat embolism. Studies in mongrel dogs after cemented arthroplasty. Am J Respir Crit Care Med 1994; 150:1416–22
Pugsley W, Klinger L, Paschalis C, Treasure T, Harrison M, Newman S: The impact of microemboli during cardiopulmonary bypass on neuropsychological functioning. Stroke 1994; 25:1393–9
Markus HS, Clifton A, Buckenham T, Brown MM: Carotid angioplasty. Detection of embolic signals during and after the procedure. Stroke 1994; 25:2403–6
Ringelstein EB, Droste DW, Babikian VL, Evans DH, Grosset DG, Kaps M, Markus HS, Russell D, Siebler M: Consensus on microembolus detection by TCD. Stroke 1998; 29:725–9
Sliwka U, Georgiadis D: Clinical correlations of Doppler microembolic signals in patients with prosthetic cardiac valves. Analysis of 580 cases. Stroke 1998; 29:140–3
Barbut D, Lo Y-W, Gold JP, Trifiletti RR, Yao FSF, Hager DN, Hinton RB, Isom OW: Impact of embolization during coronary artery bypass grafting on outcome and length of stay. Ann Thorac Surg 1997; 63:998–1002
Barbut D, Hinton RB, Szatrowski TP, Hartman GS, Bruefach M, Williams-Russo P, Charlson ME, Gold JP: Cerebral emboli detected during bypass surgery are associated with clamp removal. Stroke 1994; 25:2398–402
O’Keeffe ST, Ní Chonchubhair Á: Postoperative delirium in the elderly. Br J Anaesth 1994; 73:673–87
Williams-Russo P, Sharrock NE, Mattis S, Szatrowski TP, Charlson ME: Cognitive effects after epidural vs general anesthesia in older adults. A randomized trial. JAMA 1995; 274:44–50
Sharrock NE, Salvati EA: Hypotensive epidural anesthesia for total hip arthroplasty. Acta Orthop Scand 1996; 67:91–107
Spencer MP, Lawrence GH, Thomas GI, Sauvage LR: The use of ultrasonics in the determination of arterial aeroembolism during open-heart surgery. Ann Thorac Surg 1969; 8:489–97
Sharrock NE, Go G, Harpel PC, Ranawat CS, Sculco TP, Salvati EA: Thrombogenesis during total hip replacement. Clin Orthop 1995; 319:16–27
Shanoudy H, Soliman A, Raggi P, Liu JW, Russell DC, Jarmukli N: Prevalence of patent foramen ovale and its contribution to hypoxemia in patients with obstructive sleep apnea. Chest 1998; 113:91–6
Williams-Russo P, Urquhart BL, Sharrock NE, Charlson ME: Predictors and prognosis of post-op delirium. J Am Geriatr Soc 1992; 40:759–67