Eur J Cardiothorac Surg 2009;35:89-95. doi:10.1016/j.ejcts.2008.09.020
Copyright © 2009, European Association for Cardio-thoracic Surgery. Published by Elsevier. All rights reserved.
A method to distinguish between gaseous and solid cerebral emboli in patients with prosthetic heart valves
Rosendo A. Rodrigueza,*,
Howard J. Nathanb,
Marc Ruela,
Fraser Rubensa,
David Dafoec,
Thierry Mesanaa
a Division of Cardiac Surgery, Department of Surgery, University of Ottawa Heart Institute, Ottawa, Ontario K1Y 4W7, Canada
b Division of Cardiac Anesthesia, Department of Anesthesia, University of Ottawa Heart Institute, Ottawa, Ontario K1Y 4W7, Canada
c Department of Respiratory Therapy, University of Ottawa Heart Institute, Ottawa, Ontario K1Y 4W7, Canada
Received 6 February 2008;
received in revised form 4 September 2008;
accepted 9 September 2008.
* Corresponding author. Address: University of Ottawa Heart Institute, Division of Cardiac Surgery, Room H-4403, 40 Ruskin Street, Ottawa, Ontario K1Y 4W7, Canada. Tel.: +1 613 761 4263; fax: +1 613 761 4392. (Email: Rrodriguez{at}Ottawaheart.ca).
 |
Abstract
|
|---|
Background: The difficulty of distinguishing solid from air emboli using transcranial Doppler has limited its use in situations where both types of emboli can occur, such as in mechanical heart valve patients. To make transcranial Doppler clinically useful, a method must be found to distinguish benign air bubbles from the more damaging solid particulates. Since inhalation of 100% oxygen reduces the amount of air bubbles in mechanical heart valve patients, the ultrasonic features of the remaining emboli would be characteristic of solid particulates. Objective: We determined the accuracy of the signal relative intensity measured with transcranial Doppler to distinguish between gaseous and non-gaseous emboli in mechanical heart valve patients examined during room air and 100% oxygen. Embolic signals detected in patients with bioprosthetic valves examined during 100% oxygen comprised the source of solid particulates. Methods: Embolic signals were detected during room air (n
= 141) and 100% oxygen (n
= 45) from 17 mechanical valve patients at two Doppler examinations (4 h and 4 days after surgery). Solid embolic signals (n
= 31) from seven patients with bioprosthetic valves were identified with 100% oxygen within the first 4 h after surgery. Frequency plots and receiver operating characteristic curves assessed signal intensity differences between mechanical and bioprosthetic valve groups during 100% oxygen and the efficacy of the relative intensity for differentiating gaseous from solid emboli. Results: Administration of 100% oxygen during transcranial Doppler examination in mechanical heart valve patients decreased the count of embolic signals compared with room air (p
= 0.006). The embolic signals of mechanical heart valve patients breathing 100% oxygen showed lower relative intensities compared with those during room air. The distribution of the signal relative intensity between mechanical and bioprosthetic valve groups during 100% oxygen was similar. A 16 dB cut-off threshold achieved the best accuracy for differentiating non-gaseous from gaseous emboli (sensitivity: 60%; specificity: 82%; area: 0.721; p
< 0.0001). Conclusions: The use of a signal intensity cut-off offers adequate discrimination of the embolic composition in mechanical heart valve patients. Future studies evaluating prophylactic treatments of thrombosis in these patients should assess the predictive value of this intensity threshold and their potential association with outcome indicators and procoagulant markers.
Key Words: Heart valve prosthesis • Intracranial embolism • Ultrasonography • Doppler • Transcranial
|
1. Introduction
|
|---|
High-intensity transient signals (HITS) are the ultrasonic signatures of air or solid emboli in the brain circulation as detected by transcranial Doppler (TCD) [1]. HITS have been identified in the middle cerebral arteries of patients with mechanical heart valves (MHV) after surgery [2]. In MHV patients, most embolic signals are associated with air microbubbles that result from a process called cavitation (rapid reductions in fluid pressure) [3,4], but an unknown fraction may be represented by solid microemboli. This is in contrast to HITS detected in patients with bioprosthetic heart valves, which are considered to be non-gaseous emboli due to the absence of valve cavitation [2,3–6]. Since conventional TCD technology does not differentiate between air and solid emboli [7], the presence of air microbubbles in MHV patients confounds the identification of solid microemboli.
The inability of TCD to differentiate between gaseous and solid emboli has limited the use of this technology in situations where both embolic compositions can occur, such as in MHV patients. This issue is clinically relevant because previous studies have suggested that air emboli appear to have a minimal impact on neurologic outcome [8] while solid macro and microemboli are believed to be associated with stroke and cognitive deficits [9]. Consequently, in order to make TCD clinically useful as an outcome indicator, a method must be found to distinguish between the benign air bubbles from the more damaging solid particulates. Inhalation of 100% oxygen during TCD examination is a method that would facilitate the distinction between gaseous and non-gaseous emboli in MHV patients [10,11]. During oxygen inhalation, there is a decrease in the counts of HITS [11] due to a reduction in the amount of air bubbles [10]. We hypothesized that by reducing the number of gaseous emboli, the ultrasonic features of the remaining HITS detected with 100% oxygen would be characteristic of solid emboli. In this investigation, we compared the relative intensity [12,13] of HITS detected in MHV patients recorded during conditions of room air and 100% oxygen. In addition, we compared the relative intensity of HITS between patients with MHV and bioprosthetic valves during 100% oxygen. We postulated that the ultrasonic intensities of HITS between MHV and bioprosthetic valve patients would be similar during 100% oxygen. This would support the notion that continuous inhalation of 100% oxygen in MHV patients would improve the ability of TCD for detecting non-gaseous emboli.
|
2. Methods
|
|---|
2.1 Study population
After approval from our institutional human research ethics board and after obtaining informed consent, we enrolled patients undergoing primary mitral and/or aortic valve replacement with a mechanical heart valve (MHV group). A second group of patients (bioprosthetic valve) was studied and included patients undergoing primary aortic valve replacement with a bioprosthetic valve who were participating in another trial aimed to assess the effects of two prophylactic treatments of thrombosis on cerebral microemboli. Patients with a previous history of stroke, neurologic or psychiatric disease, history of thromboembolic events, carotid artery disease, chronic obstructive pulmonary disease (COPD), atrial fibrillation, obesity (BMI >30), allergy to any of the anticoagulants or antithrombotic drugs and carotid stenosis greater than 50% as demonstrated by carotid Duplex were excluded.
2.2 Anticoagulation in MHV
Postoperative anticoagulation included daily doses of warfarin (Coumadin®) and aspirin (ASA® 81 mg) starting the morning following surgery. Anticoagulation was supplemented with subcutaneous or intravenous heparin initiated within the first 24 h after surgery until therapeutic international normalized ratio (INR). Therapeutic INR was defined between 2.0 and 2.5 for aortic valve patients with low risk for thromboembolism or between 2.5 and 3.0 for cases at higher risk. A therapeutic INR between 2.5 and 3.5 was targeted after replacement of the mitral valve or combined aortic and mitral valves.
2.3 Anticoagulation in bioprosthetic valves
Anticoagulation included intravenous or subcutaneous heparin initiated within the first 24 h after surgery. This was followed by either one of the two prophylactic treatments of thrombosis currently used at our institution during the first 3 months after surgery: (a) daily doses of oral aspirin (81 mg) and warfarin with a therapeutic INR between 2.0 and 2.5; or (b) daily doses of aspirin (325 mg).
2.4 Clinical follow-up
Patients were followed during their stay in hospital looking for clinical complications. In all patients, length of stay in hospital, type of anticoagulation regimen, and daily values of procoagulant markers such as the INR, platelet counts and partial thromboplastin time (PTT) were documented.
2.5 TCD in MHV patients
MHV patients were examined after surgery using a dual-gated pulsed-wave TCD system (MDT2, DWL, Sipplingen, Germany) equipped with two 2 MHz probes. An initial TCD examination was performed within the first 4 h after the surgery (examination 1) while the patient was still intubated and under ventilatory support. Subsequently, patients were re-examined 4 days after surgery (examination 2) while being awake and at their bedside. Prior to TCD monitoring, a vascular Doppler examination on each temporal window was performed with marking of the location of the best Doppler signal corresponding to the main trunk (M-1 segment) of the middle cerebral artery (MCA) [1]. An adjustable headband (Marc 600, Spencer Technologies, Seattle, WA) secured the TCD probes on the temporal windows for bilateral monitoring of both MCA. TCD recordings were performed using the following parameters: 58% overlapping, FFT resolution of 128 Hz, dynamic range of 60 dB, an intensity threshold greater than 7 dB, inter-gate separation of 5 mm, sample volume of 10 mm and high-pass filtering with a cut-off frequency of 100 Hz.
2.6 TCD in bioprosthetic valve patients
Patients were examined within the first 4 h after surgery and HITS were monitored for 30 min with 100% oxygen while intubated and under ventilatory support. TCD instrumentation included a power M-mode digital (PMD-100) TCD (Spencer Technologies, Seattle, WA) equipped with bilateral 2 MHz Doppler probes and a recording system (software version 1.3.6; Spencer Technologies, Seattle WA) with 33 gates placed at 2 mm intervals along the ultrasonic beam (sample volume: 6 mm; M-mode display range: 22–87 mm; digital filter: 150 Hz) [14]. The vascular examination protocol and monitoring procedures were similar to the MHV group with both MCA gates selected for spectral analysis. Doppler spectrogram and power mode data were stored on hard disk and replayed during the off-line analysis.
2.7 Breathing conditions
TCD examination in MHV patients alternated between 30 min with room air and 30 min with 100% oxygen. This sequence was randomized using computer generated random numbers. If awake, patients were asked to breath normally under conditions of room air, but if the patients were still intubated, standard ventilatory settings were used including positive end-expiratory pressure (PEEP) of 5 cm of water (H2O) and fraction inspired of oxygen (FiO2) of 30%, to maintain systemic arterial oxygen saturations greater than 95%. In addition, partial pressures of carbon dioxide (CO2) were maintained within normal values (35–45 mmHg) with tidal volumes of 6–8 ml/kg and respiratory rates of 10–14 breaths per min.
During inhalation of oxygen, TCD recordings were initiated 3 min after the beginning of continuous 100% oxygen flow. The administration of 100% oxygen in intubated patients was achieved with the same ventilator parameters than the room air condition except that the FiO2 was set to deliver 100% oxygen. In awake and extubated patients, 100% oxygen was administered through the use of a BiPAP VISION® ventilatory support system (Respironics, Inc. Murrysville, PA). Patients were breathing spontaneously through a NIV clear facial mask. The mask was placed over the mouth and nose and held in place using an appropriate headgear to ensure a tight seal. The ventilator was set to apply a continuous positive airway pressure (CPAP) of 5 cm of H2O and oxygen concentration of 100%. Patients were instructed to breath normally, avoid hyperventilation or hypoventilation and immediately give notice if breathing became uncomfortable or other cardiorespiratory complaints occurred. Leakage of the mask was immediately detected by a decrease in the level of the pre-established CPAP and it was immediately corrected. A staff respiratory therapist always assisted the examiner (RAR) during this phase of the investigation.
2.8 TCD data analysis
An experienced ultrasonographer (RAR) reviewed the Doppler recordings from both groups at least 1 week after TCD examination. Doppler signals were classified as true HITS, equivocal HITS, artifacts and Doppler speckles according to pre-established criteria [15,16]. Equivocal HITS, artifacts and Doppler speckles were not included in the final analysis. Therefore, only true HITS were considered in the final counts of HITS and they were classified according to whether these signals were detected under 100% oxygen or room air.
In MHV patients, the relative intensity of true HITS was measured using the automated software of the MDT2 system (Multi-flow, MF version 8.27j, DWL, Inc, Sipplingen, Germany). The software calculates the relative intensity as the ratio between the power of the embolic signal and that from the background blood over the whole spectral screen where no embolic signal was present [15]. Relative intensity was then calculated as: relative intensity = 10 log [power of embolic signal/background power].
In bioprosthetic valve patients, true HITS were selected if their presence was simultaneously identified in both the Doppler and power modes of the power M-mode digital TCD system. Since the PMD100 does not include automatic software for calculating the signal intensity, the relative intensity of HITS was calculated manually [17] using the color-coded intensity scale of the Doppler mode expressed in decibels (dB). For these purposes, the color-coded intensity of the Doppler spectra was decreased through increases in the noise threshold (this was equivalent to a reduction in the gain of the system) until the color of the Doppler spectrum corresponding to the red blood cells in the background blood from the cardiac cycle adjacent to the embolic signal changed to black. The decibel value on the color-coded intensity-scale corresponding to this level of noise was designated as the intensity for the background blood. Subsequently, we continued increasing the noise threshold until the color of the spectrum corresponding to the embolic signal changed to black and the new decibel value on the color-coded intensity scale was designated as the intensity of the HITS. The difference between these two intensity values (HITS minus background blood) expressed in decibels (dB) was used as the manually calculated relative intensity of HITS.
2.9 Statistical analysis
Continuous variables are reported as the means and their 95% confidence intervals (95% CI). Paired differences in the counts of HITS between breathing conditions or TCD examinations were assessed by Wilcoxon signed rank tests. Frequency histograms were used to display differences in the distribution of the relative intensity of HITS according to experimental conditions and valve groups. The distribution of the relative intensity of HITS in bioprosthetic valve patients breathing 100% oxygen was considered as indicative of the characteristics of solid microemboli. Receiver operating characteristic (ROC) curves determined the efficacy of the relative intensity for differentiating between HITS detected during 100% oxygen compared with those during room air. The cut-off point that would reject the highest proportion of HITS during room air (sensitivity) while detecting the largest counts of HITS during 100% oxygen (specificity) was defined as the optimal threshold (best cut-off point). All tests used p
< 0.05 as the critical value of statistical significance. Analyses were performed using the SPSS package (version 15.0).
|
3. Results
|
|---|
3.1 Patient population
3.1.1 Mechanical heart valve group
Seventeen patients were studied in this group. All their native valves were replaced with ON-X® valves (Edwards Lifesciences, Palo Alto, CA). In 12 patients the aortic valve was replaced, in 3 additional patients both the aortic and mitral valves and in 2 cases only the mitral valve was exchanged. The average time from arrival to the intensive care unit (ICU) to the first TCD examination was 137 min (95% CI: 122, 152 min) and to the second assessment 4 days (95% CI: 3, 4 days). All patients except one completed both TCD evaluations. One patient finished his oxygen session by using a conventional partial non-re-breathing mask (100% oxygen, flow: 15 l/min) due to claustrophobia. There were no manifestations of numbness, dizziness or seizures attributed to the inhalation of 100% oxygen. All patients were discharged without any clinical deficits.
3.1.2 Bioprosthetic valve group
Seven patients (see Table 1
) were evaluated in this group. Three patients had their aortic valve replaced with a perimount Carpentier Edwards porcine bioprosthesis (Edwards Lifesciences Inc, Mississauga, ON), another three with a Hancock II valve (Medtronic Inc. Minneapolis, MN) and one with a Medtronic Mosaic bioprosthesis (Medtronic Inc, Minneapolis, MN). The average time from ICU arrival to TCD assessment for this group of patients was 148 min (95% CI: 130, 167 min) (range: 121–182 min). None of these patients showed any in-hospital complication.
View this table:
[in this window]
[in a new window]
|
Table 1 Patients’ age, HITS counts and INR values between the two valve groups. Values are expressed as their means (95% confidence interval limits).
|
|
3.2 HITS counts
HITS were detected in all MHV patients in at least one of the two TCD examinations. All patients with bioprosthetic porcine valves except one, showed HITS immediately after surgery. Table 1 shows the mean counts of HITS in both valve groups of patients according to the breathing condition, TCD examination and INR values. In MHV patients, the mean counts of HITS recorded during room air tended to increase from the first to the second TCD examination (p
= 0.104), but the counts were similar between the two TCD assessments during 100% oxygen (p
= 0.326). Continuous inhalation of 100% oxygen in MHV patients significantly decreased the counts of HITS compared with room air at the second TCD assessment (see Table 1; p
= 0.006). However, there was no significant difference in the counts of HITS between breathing conditions at the first TCD evaluation after surgery (p
= 0.121) possibly because of the small number of HITS at the initial assessment. During room air, HITS were detected in 53% of MHV patients within the first few hours after surgery compared with 75% 4 days after (p
= 0.141). The total sum of HITS in all MHV patients including both examinations at room air was 141 HITS (45% on the left MCA). This amount decreased to 45 HITS (44% on the left MCA) in the same group during 100% oxygen and to 31 HITS (45% on the left MCA) in the bioprosthetic valve group. A single patient in the MHV group was examined with TCD while in atrial fibrillation. The rate of HITS for this patient was the highest during the administration of 100% oxygen.
3.3 Intensity characteristics of HITS
Fig. 1
illustrates the frequency distributions of the relative intensities for all HITS according to their valve group and breathing condition. The frequency histogram of the HITS relative intensities in MHV patients breathing room air showed a tri-modal distribution where 37% of HITS had signal intensities between 8 and 15 dB (mode 10 dB), 29.8% between 16 and 21 dB (mode: 18 dB) and 33.2% between 22 and 32 dB (mode: 25 dB). In contrast, HITS detected during 100% oxygen in MHV patients displayed a bimodal distribution toward the lower signal intensities compared with those at room air. With 100% oxygen, 84.2% of HITS in MHV patients had signal intensities between 8 and 16 dB (mode: 12 dB) and only 15.8% between 17 and 26 dB (mode: 19 dB). This frequency distribution of HITS was comparable to the bioprosthetic valve group breathing 100% oxygen where 80.8% of HITS had relative intensities between 6 and 16 dB (mode: 12 dB) while only 19.2% displayed intensities between 17 and 20 dB (mode: 19 dB). Fig. 2
illustrates the distributions of the HITS intensity according to the two conditions (room air and 100% oxygen) in four MHV patients.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 1. Frequency distribution (vertical-axis) of the HITS relative intensities (horizontal-axis) in patients with mechanical heart valves (MHV) and bioprosthetic valves (bioprosthesis). The MHV group was tested at room air (white diamonds + interrupted line) and 100% oxygen (dark squares + continuous line) while the bioprosthetic valve group only during 100% oxygen (dark circles + continuous line). Points represent the percentage of HITS within each intensity interval relative to the total. Each intensity interval represents the sum of HITS from two intensity levels. Relative intensities of HITS in MHV patients with 100% oxygen were lower than those with room air, but there were no differences between MHV and bioprosthetic valves during 100% oxygen.
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 2. HITS relative intensities (horizontal-axis) in four mechanical heart valve patients during room air (white diamonds + interrupted lines) and 100% oxygen (dark squares and continuous lines). HITS of high, medium and low intensity are found during room air. The inhalation of 100% oxygen shifts the HITS intensity distribution toward the area of lower intensity signals.
|
|
3.4 ROC curves
Fig. 3
shows the ROC curve for the signal relative intensity and the cut-off point that best separated HITS measured in MHV patients during inhalation of air versus oxygen. The ROC curve analysis indicated that 16 dB was the best cut-off threshold for differentiating HITS between the two conditions (sensitivity: 62%; specificity: 82%; area under the ROC curve: 0.721; p
< 0.0001). Fig. 4
represents the cumulative frequency distribution of the HITS relative intensity for both valve groups stratified by breathing condition. Based on the signal relative intensity, using 16 dB as the cut-off point in MHV patients, allowed differentiating 84% of HITS detected during 100% oxygen (81% in the bioprosthetic valve group) from 60% of HITS identified during room air.
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 3. Receiver operating characteristic curve showing sensitivity (vertical-axis) and specificity of the relative intensity for differentiating HITS between 100% oxygen and room air. The best discriminating cut-off is shown (arrow).
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 4. Cumulative frequency distribution of the HITS relative intensity (horizontal axis) in patients with mechanical heart valves (MHV) and bioprosthetic valves (bioprosthesis). A cut-off threshold of 16 dB (vertical arrow) separated 84% of HITS in MHV patients and 80.8% of HITS in bioprosthetic valve patients during 100% oxygen from 60.4% of HITS detected during room air.
|
|
|
4. Discussion
|
|---|
Our findings indicate that HITS in MHV patients breathing 100% oxygen have lower relative intensities than those identified during room air and likely represent solid emboli. The two types of emboli in these patients can be discriminated using a signal relative intensity cut-off of 16 dB. Also, our observations suggest that HITS recorded at room air fit into three categories of lower, medium and higher intensity signals (Fig. 1) and that the administration of 100% oxygen removes the high-intensity HITS. Since air microbubbles have greater ultrasonic echogenicity [13], HITS associated with air bubbles display higher relative intensities [12,13,18] compared with those representing non-gaseous emboli. We suggest that the change in the intensity characteristics of HITS from room air to 100% oxygen reflects a reduced number of air microbubbles. Our findings are supported by previous studies, which have documented that most HITS in MHV patients detected under normal breathing conditions are microbubbles and that inhalation of oxygen makes them disappear from the circulation [10,11,19]. Additionally, our results indicate that the intensity characteristics of HITS in MHV patients are comparable to those in bioprosthetic valves when both groups receive 100% oxygen. As bioprosthetic valves do not demonstrate cavitation, HITS detected in this group of patients are assumed to be non-gaseous emboli possibly due to the interaction between the blood elements and the biological valve [2–6]. Therefore, most HITS in MHV patients during 100% oxygen are likely non-gaseous emboli.
The use of the 16 dB cut-off derived from the ROC curve analysis on the intensity characteristics of HITS detected at room air (Fig. 3) suggested that only 39.6% of those signals are non-gaseous. This threshold is very close to our 18 dB cut-off found in our previous in vitro study [18] where solid spheres were differentiated from air bubbles of similar size in an ultrasound phantom-loop using a 2 MHz Doppler-transducer. In comparing both of our studies, 100% of HITS associated with solid spheres in our in vitro model, all HITS from bioprosthetic valve recipients and 94% of those from MHV patients during 100% oxygen displayed signal relative intensities equal or less than 19 dB (Fig. 4). In contrast, 82% of HITS corresponding to air bubbles injected in vitro showed relative intensities greater than 17 dB while only a small fraction (18%) had intensity characteristics between 15 and 17 dB. This distribution is similar to the 60.4% of HITS found in MHV patients during room air with relative intensities above the 16 dB cut-off (which made them classified as gaseous emboli).
Previous studies [13] have suggested that using a simple cut-off threshold based on the signal intensity may be impractical clinically because of the confounding effects of the embolus size. The problem is that signal intensity increases with embolus size [13] and if neither size nor composition is known, accurate classification may be difficult. Recently, Markus and Punter [7] compared the accuracy of classifying solid and air emboli by using a simple intensity cut-off (range: 5–20 dB) obtained with a single transducer (2 MHz) against another derived from the intensity differences at two different frequency transducers (2.0 and 2.5 MHz). They used as a source of solid emboli a group of patients with symptomatic carotid stenosis and for air bubbles, the HITS recorded during injection of agitated saline in patients with patent foramen ovale. At a fixed sensitivity of 50%, their specificity for discriminating solid particulate from air bubbles was 94% with the single frequency transducer and 96% with the dual. Their results would be comparable to our current study by shifting the cut-off threshold from 16 to 19 dB. Then, sensitivities would be comparable (50%) and our specificity would increase from 82% to 86%.
HITS counts in our patients during room air were lower than those observed in previous studies [10,11]. We speculate that lower HITS counts may be a reflection of the type of mechanical valve used in our study (ON-X valve) as differences in the design and geometry of the valves can influence the amount of cavitation [4].
Nitrogen washout from the lungs due to the administration of 100% oxygen occurs exponentially. After 3 min of breathing 100% oxygen, only 2.1 ± 0.7% nitrogen remains in the alveoli [20]. This is in contrast to blood where denitrogenation occurs rapidly during the initial phase [21], but becomes slow at a later stage and a complete denitrogenation could take up to 3 h [11]. Although it was assumed that continuous oxygenation eliminated a large proportion of air microbubbles from the circulation, a complete denitrogenation is difficult to prove. Our findings show that a small proportion (4.6%) of HITS in MHV patients breathing 100% oxygen showed intensities in the range between 22 and 26 dB compared with none in patients with bioprosthetic valves (Fig. 1). This small proportion of HITS with higher relative intensities during 100% oxygen may be consistent with either small microbubbles or larger size solid particulates. The fact that none of our patients showed complications during oxygen administration suggests that this method is safe.
|
5. Conclusion
|
|---|
In summary, administration of 100% oxygen in MHV patients decreases the count of HITS compared with room air. The HITS of MHV patients breathing 100% oxygen show lower relative intensities compared with those during room air. Additionally, the relative intensities of HITS between MHV and bioprosthetic valve patients detected with 100% oxygen are comparable. Our current findings support the conclusion that the majority of cerebral emboli detected with TCD during oxygen inhalation in MHV patients are non-gaseous. Our results indicate that 16 dB is the best cut-off that would differentiate the populations of HITS between the two breathing conditions. The application of this threshold to HITS detected at room air in MHV patients shows that 60% of higher intensity embolic signals are excluded due to the inhalation of oxygen and that these signals are likely of gaseous composition. The remaining fraction (40%) of lower intensity signals which are not affected by oxygen and whose intensity characteristics are similar to those in bioprosthetic valve patients likely corresponds to solid emboli. Future clinical studies evaluating prophylactic treatments of thrombosis in MHV patients should assess the predictive value of this intensity threshold for differentiating between these two populations of HITS and their potential association with outcome indicators and procoagulant markers.
|
Acknowledgments
|
|---|
We thank the staff of the University of Ottawa Heart Institute for their co-operation.
|
References
|
|---|
- Ringelstein EB, Droste DW, Babikian VL, Evans DH, Grosset DG, Kaps M, Markus HS, Russell D, Siebler M. Consensus on microembolus detection by TCD. International consensus group on microembolus detection. Stroke 1998;29:725-729.[Abstract/Free Full Text]
- Geiser T, Sturzenegger M, Genewein U, Haeberli A, Beer JH. Mechanisms of cerebrovascular events as assessed by procoagulant activity, cerebral microemboli, and platelet microparticles in patients with prosthetic heart valves. Stroke 1998;29:1770-1777.[Abstract/Free Full Text]
- Deklunder G, Lecroart JL, Savoye C, Coquet B, Houdas Y. Transcranial high-intensity Doppler signals in patients with mechanical heart valve prostheses: their relationship with abnormal intracavitary echoes. J Heart Valve Dis 1996;5:662-667.[Medline]
- Johansen P. Mechanical heart valve cavitation. Expert Rev Med Devices 2004;1:95-104.[CrossRef][Medline]
- Sliwka U, Georgiadis D. Clinical correlations of Doppler microembolic signals in patients with prosthetic cardiac valves: analysis of 580 cases. Stroke 1998;29:140-143.[Abstract/Free Full Text]
- Dexter EU, Aluri S, Radcliffe RR, Zhu H, Carlson DD, Heilman TE, Chandran KB, Richenbacher WE. In vivo demonstration of cavitation potential of a mechanical heart valve. ASAIO J 1999;45:436-441.[Medline]
- Markus HS, Punter M. Can transcranial Doppler discriminate between solid and gaseous microemboli? Assessment of a dual-frequency transducer system. Stroke 2005;36:1731-1734.[Abstract/Free Full Text]
- Neville MJ, Butterworth J, James RL, Hammon JW, Stump DA. Similar neurobehavioral outcome after valve or coronary artery operations despite differing carotid embolic counts. J Thorac Cardiovasc Surg 2001;121:125-136.[CrossRef][Medline]
- Hindman BJ. Emboli, inflammation and CNS impairment: an overview. Heart Surg Forum 2002;5:249-253.[Medline]
- Kaps M, Hansen J, Weiher M, Tiffert K, Kayser I, Droste DW. Clinically silent microemboli in patients with artificial prosthetic aortic valves are predominantly gaseous and not solid. Stroke 1997;28:322-325.[Abstract/Free Full Text]
- Georgiadis D, Wenzel A, Lehmann D, Lindner A, Zerkowski HR, Zierz S, Spencer MP. Influence of oxygen ventilation on Doppler microemboli signals in patients with artificial heart valves. Stroke 1997;28:2189-2194.[Abstract/Free Full Text]
- Smith JL, Evans DH, Bell PR, Naylor AR. A comparison of four methods for distinguishing Doppler signals from gaseous and particulate emboli. Stroke 1998;29:1133-1138.[Abstract/Free Full Text]
- Markus HS, Brown MM. Differentiation between different pathological cerebral embolic materials using transcranial Doppler in an in-vitro model. Stroke 1993;24:1-5.[Abstract/Free Full Text]
- Moehring MA, Spencer MP. Power M-mode Doppler (PMD) for observing cerebral blood flow and tracking emboli. Ultrasound Med Biol 2002;28:49-57.[CrossRef][Medline]
- Rodriguez RA, Rubens F, Rodriguez CD, Nathan HJ. Sources of variability in the detection of cerebral emboli with transcranial Doppler during cardiac surgery. J Neuroimaging 2006;16:126-132.[Medline]
- Rodriguez RA, Williams KA, Babaev A, Rubens F, Nathan HJ. Effect of perfusionist technique on cerebral embolization during cardiopulmonary bypass. Perfusion 2005;20:3-10.[Abstract/Free Full Text]
- Rodriguez RA, Giachino A, Hosking M, Nathan HJ. Transcranial Doppler characteristics of different embolic materials during in vivo testing. J Neuroimaging 2002;12:259-266.[CrossRef][Medline]
- Rodriguez RA, Rodriguez CD, Mesana T, Nathan HJ. Distinguishing air from solid emboli using ultrasound: in-vitro study of the effect of Doppler carrier frequency. J Neuroimaging 2007;17:211-218.[Medline]
- Baumgartner RW, Frick A, Kremer C, Oechslin E, Russi E, Turina J, Georgiadis D. Microembolic signal counts increase during hyperbaric exposure in patients with prosthetic heart valves. J Thorac Cardiovasc Surg 2001;122:1142-1146.[Abstract/Free Full Text]
- Carmichael FJ, Cruise CJE, Crago RR, Paluk S. Preoxygenation: a study of denitrogenation. Anesth Analg 1989;68:406-409.[Free Full Text]
- Finley TN. The determination of uneven pulmonary blood flow from the arterial oxygen tension during nitrogen washout. J Clin Invest 1961;40:1727-1734.[CrossRef][Medline]
Top
Abstract
1. Introduction
