| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;78:875-881
© 2004 The Society of Thoracic Surgeons
a Department of Cardiology, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands
b Department of Intensive Care Medicine, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands
c Laser Center, Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands
d Department of Nuclear Medicine, Academic Medical Center, University of Amsterdam, Amsterdam The Netherlands
e Department of Cardiothoracic Surgery, Academic Medical Center, University of Amsterdam, Amsterdam, the Netherlands
Accepted for publication February 18, 2004.
* Address reprint requests to Dr Beek, Laser Center, Academic Medical Center, University of Amsterdam, Meiberg dreef 9, 1105AZ Amsterdam, PO Box, 22700 1100DE Amsterdam, the Netherlands
j.f.beek{at}amc.uva.nl
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: Thirty patients with refractory angina were randomized in pairs to excimer TMLR or maximal medication. We assessed angina, quality of life (QOL), exercise time, myocardial perfusion, and ventricular wall motion at base line and at 3, 6, and 12 months after TMLR.
RESULTS: TMLR patients manifested a significantly better outcome with respect to angina class and quality of life. One TMLR patient died perioperatively versus none in the control group. After TMLR angina decreased from class 3.8 ± 0.4 at base line to 1.9 ± 0.9 at 12 months versus 3.9 ± 0.3 to 3.7 ± 0.6 in the control group, respectively (p = 0.000001). At 12 months a decrease of greater than or equal to two angina classes was indicated in 11 out of 14 TMLR patients versus none in the control group (p = 0.00001). Improved myocardial perfusion or exercise time was not indicated despite a small decrease in reversible wall motion abnormality score.
CONCLUSIONS: Excimer TMLR significantly relieves angina and improves QOL without evidence of improved cardiac perfusion or function.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
Randomization
This study was designed as a randomized single-center trial consisting of 30 patients. When 2 subsequent patients met the criteria they were included as a pair, randomized between excimer TMLR and continued maximal medication, and then followed simultaneously for 1 year. At randomization patients assigned to continued maximal medication were offered to undergo TMLR after 1 year follow-up (no earlier). The study was approved by the hospital ethical committee and all patients gave written and informed consent.
TMLR procedure and medication
Premedication was identical to that prescribed for open-heart surgery patients. Under general anesthesia a left lateral thoracotomy was performed. The pericardium was opened and without cardiopulmonary bypass transmyocardial channels were created in the ischemic area of the left ventricular wall as assessed by perfusion scintigraphy. Approximately one transmyocardial channel per cm2 was created with a XeCl excimer laser (Max-20; Medolas, Munich, Germany; wavelength = 308 nm, 32–40 mJ/pulse, pulse duration = 110 ns, pulse frequency = 40 Hz) combined with a hospital-built electrocardiogram (ECG)-trigger device. The myocardium was perforated by gently and manually advancing a 1-mm-diameter flat-tipped fiber during 3–4 triggered cardiac cycles (4–5 pulses per cycle) under transesophageal echocardiographic guidance [11]. Manual pressure was usually sufficient to stop the channel opening from bleeding. Positive inotropics were avoided if possible and preoperative medication was resumed within 24 hours. Further care was similar to that given after open-heart surgery. In all patients maximal medication, defined as maximally tolerable doses of ß-blockers, Ca-antagonists, and nitrates, was continued.
Follow-Up and measurements
All pairs of patients were examined at the outpatient clinic at base line and at 1, 3, 6, and 12 months after the date of TMLR in one of them.
The functional angina status was assessed by one cardiologist (JAPvdS) using the NYHA classification. Changes in medication were recorded as well as major adverse cardiac events including hospital admittance, unstable angina, myocardial infarction, heart failure, arrhythmias, and death. Myocardial infarction was defined as new Q-waves on the ECG. Postoperatively MB-isoenzyme fraction of creatine phosphokinase (CK-MB) washout curves were determined.
QOL was assessed during follow-up visits using the Seattle Angina Questionnaire and the EuroQol Standardized Questionnaire [12, 13]. Exercise tolerance was measured using a (symptom-limited) treadmill test according to a modified Bruce protocol. Medication was continued during the test. Exercise time and the reason for stopping were recorded.
Myocardial perfusion scintigraphy was performed (at base line and at 3 and 12 months) according to the guidelines of the American Society of Nuclear Cardiology [14] using 99mtechnetium-labeled tetrofosmin (99mTc-TF) and a 2-day stress/rest protocol. Stress was induced by symptom-limited exercise, or pharmacologically for patients with a left bundle branch block and for patients in whom exercise was not possible or suboptimal. In each patient the stress modality used at base line was also used at follow-up. 99mTc-TF (approximately 500 MBq) was injected at maximal exercise or after pharmacological stress and at rest during rest testing. Single-photon emission computed tomography (SPECT) was performed 1 hour after 99mTc-TF injection. Images were obtained with the patient in prone position using a three-headed gamma camera (MultiSPECT-3; Siemens, Hoffman Estate, IL) equipped with low-energy high-resolution collimators and a window centered on the [99mTc] 140 keV photopeak according to a step-and-shoot protocol allowing 45 seconds per view. Images were reconstructed using standard filtered back projection without applying attenuation correction. Short axis slices were obtained and an 18-segment bulls eye was reconstructed. Stress and rest images were scored using a three-dimensional sampling and analysis algorithm that generates a five-point myocardial perfusion score for each myocardial segment [15]. Perfusion was classified as normal [0], equivocal abnormal [1], mildly abnormal [2], moderately abnormal [3], or severely abnormal [4]. The nuclear medicine physicians (HJV, BLFvE-S), blinded to the treatment and date of investigation, reviewed the generated scores per scintigram and segment. In case of artifact scoring by the algorithm program scores were manually adjusted. Per patient and follow-up moment a summed stress score, summed rest score, and summed difference score (generated from the summed stress score and summed rest score) was calculated.
Stress echocardiography was performed using a HDI 3000 echocardiograph (Advanced Technologies Laboratories, Bothell, WA) or a System V echocardiograph (GE Ultrasound Medical Systems, Milwaukee, WI). Images were obtained at base line and with increasing dobutamine doses (every 3 minutes, 10, 20, 30, 40, and 50 µg · kg–1 · min–1) until the target heart rate (85% of the gender and age-predicted rate) was achieved or new or increasing wall motion abnormalities, severe angina, or ischemic ECG changes were observed. The ischemic area was qualitatively described by scoring the number of involved coronary territories (0, 1, 2, or 3) for both reversible and fixed wall motion abnormalities. Stress echocardiograms were defined as nondiagnostic if the target heart rate was not reached (without another criterium to stop the test) and if echo windows were inadequate for image interpretation. The test was regarded as positive if worsening wall motion or a biphasic response of already diminished contracting myocardium was observed.
LVEF was assessed (at base line and at 12 months) using equilibrium radionuclide angiography according to the guidelines of the American Society of Nuclear Cardiology [14]. Ambulant 24-hour ECG was assessed using a Medilog Excel 2 and MR45 recorder (Oxford Instruments, Abingdon, Oxon, United Kingdom). The occurrence of ventricular arrhythmias (> five premature ventricular contractions per minute and of ventricular tachycardia) was recorded. Episodes of angina were recorded in a patient diary.
End points and statistical analysis
The primary end point was the improvement of angina at the 12-month follow-up after TMLR compared with the control group. Secondary end points were medical treatment, clinical events, QOL, exercise time, and myocardial perfusion and function.
Therapeutic efficacy was defined as an improvement in angina of greater than or equal to 2 classes in 40% of the TMLR patients compared with no improvement in the control group (
= 5%, 1-ß = 80%). Loss to follow-up was expected to be less than 10% resulting in negligible loss of power.
In both groups missing QOL scores and exercise times (in between base line and the 12-month follow-up) were substituted using linear interpolation. When the first scheduled postoperative follow-up was missing in the TMLR group (eg, 1 month QOL or 3 month exercise data), the average (TMLR) group changes were used to calculate substitutes for the missing values. Deceased patients were included in the analysis of base line characteristics (Table 1) but excluded from the overall results. In a subanalysis (of angina, visual analogue scale of the EuroQol questionnaire and exercise time) deceased patients were taken into account by including their base line values at 1, 3, 6, and 12 months (last observation carried forward; Table 2). If no diagnostic stress echocardiogram was available at base line or at 12 months, the results at 3 or 6 months were extrapolated if no relevant clinical events had occurred.
|
|
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
Surgical and perioperative data
During TMLR 46 ± 10 channels were made. Transesophageal echocardiography confirmed transmyocardial perforation in 88% of 696 channels. One channel required epicardial suturing to stop the bleeding. Skin-to-skin and TMLR-procedure times were 111 ± 22 and 24 ± 11 minutes, respectively. No clinically important arrhythmias or bleedings were observed. Postoperative myocardial infarctions (CK-MBmax values of 101, 108, 221, and 697 µg/l, respectively) developed in 4 patients. The infarction developed during the operation in 1 patient (CK-MBmax = 697 µg/l) and this patient died the same day (perioperative and total 1-year mortality of 6.7% in the TMLR group). The 11 patients without perioperative myocardial infarction all demonstrated CK-MBmax values lower than 45 µg/l (mean 27 ± 9 µg/l). Postoperative pneumonia developed in 2 of the 14 surviving patients. In 1 patient 21 days of mechanical ventilation was necessary resulting in a 51 days in-hospital stay. The remaining 13 patients had 22 ± 9 hours mechanical ventilation and 9 ± 1 days in-hospital stay.
Angina class
Changes in the NYHA angina class are illustrated in Table 2 and Figure 1. At the 12-month follow-up 11 out of 14 TMLR patients (79%) improved greater than or equal to two classes in angina versus 0 out of 15 control patients. Angina was already significantly improved at 1-month after TMLR. The improvement in angina class at 12 months after TMLR was significantly better than in the control group (from class 3.8 ± 0.4 at base line to 1.9 ± 0.9 at 12 months after TMLR vs 3.9 ± 0.3 to 3.7 ± 0.6, respectively, in the control group; p = 0.000001, Table 2). Even when the base line value of the deceased patient was included (angina class IV/IV during follow-up) statistical significance remained.
|
Quality of life
The scores (%) of the physical limitation domain, angina frequency domain, and disease perception domain of the Seattle Angina Questionnaire improved significantly (p = 0.001, p = 0.09, and p = 0.09, respectively) after TMLR compared with the control group (the TMLR group respectively from 40.5 ± 12.9, 27.1 ± 24.0, and 32.7 ± 20.5 at base line to 59.3 ± 13.7, 52.1 ± 30.4, and 63.7 ± 18.7 at 12 months vs the control group respectively from 32.8 ± 15.5, 30.7 ± 26.6, and 20.0 ± 14.0 at base line to 30.9 ± 19.1, 26.7 ± 29.7, and 27.2 ± 24.3 at 12 months). Compared with base line the score of the anginal stability domain at 12 months was also significantly improved after TMLR, but the score for the treatment satisfaction domain was not improved. The EuroQol questionnaire's visual analogue scale increased significantly after TMLR compared with the control group (p = 0.002; Table 2). When we included the base line score (40%) of the deceased patient during follow-up statistical significance remained.
Exercise time
Change in exercise time from base line to the 12-month follow-up was not notably different in the two groups (Table 2). However at 12 months after TMLR the incidence of angina as a reason for terminating the exercise test was significantly decreased compared with preoperatively whereas this was unchanged in the control group.
Myocardial perfusion scintigraphy
First, all postoperative myocardial perfusion scans showed sustained ischemia. In the TMLR group the mean summed difference score at base line was 13.9 ± 7.8 versus 12.1 ± 7.8 and 11.7 ± 5.2 at 3 and 12 months, respectively (no considerable differences compared with base line). In the control group the mean summed difference score at base line was 10.9 ± 5.7, 9.5 ± 6.3, and 9.4 ± 7.4, respectively (no considerable differences compared with base line and compared with the TMLR group). Second, the summed rest scores at 12 months did not differ from those at base line (both within and between groups).
Stress echocardiography
The reversible wall motion abnormality score was significantly decreased at 12 months after TMLR compared with the control group (from 1.1 ± 0.5 at base line to 0.5 ± 0.5 at 12 months after TMLR vs 1.1 ± 0.6 and 1.2 ± 0.8, respectively, in the control group; p = 0.005) whereas the fixed wall motion abnormality score was significantly increased (from 0.3 ± 0.5 at base line to 0.7 ± 0.5 at 12 months after TMLR vs 0.3 ± 0.5 and 0.5 ± 0.7, respectively, in the control group; p = 0.008).
Ejection fraction
Average preoperative LVEF was 55% ± 9% in the TMLR group and 56% ± 6% in the control group (p = 0.67). Compared with base line the LVEF was decreased significantly at 12 months after TMLR (51% ± 11%, p = 0.01). In the control group LVEF was not significantly decreased (p = 0.07) at 12 months (53% ± 8%).
Ambulant 24-hour ECG
Compared with base line the reported incidence of angina was significantly decreased at 12 months after TMLR (p = 0.003) as opposed to a decrease that was not significant in the control group. However the comparison of the change in anginal class between the groups did not reach statistical significance. Compared with the control group the incidence of premature ventricular contractions was unchanged at 12 months after TMLR as was the incidence of ventricular tachycardia (ventricular tachycardia was not observed in any patients).
|
Comment |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
Clinical improvement has been reported after excimer TMLR [8, 9], however we have performed a randomized study. The results of this study indicate that at 12-months follow-up, excimer TMLR significantly reduces angina pectoris and improves QOL compared with maximal medication.
Angina, QOL, and exercise time
In the current study 79% of the patients in the TMLR group presented a decrease of greater than or equal to two angina classes at the 12-month follow-up versus 0% of the patients in the control group. The ambulant ECG diary confirmed a significant relief of angina at 12 months after TMLR. Although assessed in a smaller number of patients this result is (at least) as favorable as the outcome of earlier randomized TMLR studies using CO2 and holmium:YAG lasers. In these studies angina decreased with greater than or equal to two classes in 24%–76% of the TMLR patients and with greater than or equal to one class in at least 63% of the TMLR patients (vs 86% in this study). A clinically important observation in both earlier randomized trials as well as in our study is that the therapeutic gain is highest in the patients with the most severe angina. In the current study 80% of the TMLR patients had NYHA class IV at base line and 79% of all TMLR patients (including those with NYHA class III) improved greater than or equal to two classes at 12 months. Furthermore 91% of the improved patients were in class IV at base line. In the other studies base line class IV percentages and the percentages of all patients that improved by greater than or equal to two classes at 12 months were respectively 100% and 76% [5], 63% and 61% [3], 70% and 72% [1, 4], 27% and 24% [2], and 24% and 39% [6]. These numbers indicate that the benefit of TMLR regarding angina is the most pronounced in the patients with the most severe angina. The results of our study illustrate that in patients with refractory angina, excimer TMLR is at least as effective in reducing angina as CO2 and holmium:YAG TMLR.
The improvement regarding QOL in our study is comparable with results reported earlier by Spertus and associates [7] and Allen and associates [5]. As in our opinion improvement in QOL is the major goal of treatment regarding refractory angina, this remarkable improvement is clinically even more important than the observed improvement in angina class.
Given the results indicated above we surprisingly did not find any evidence for increased exercise time. Three [3, 5, 6] out of four [2, 3, 5, 6] randomized studies that assessed exercise time did report substantial improvements in the TMLR group. It is unlikely that this can be attributed to the use of excimer TMLR. An explanation might be the use of a modified Bruce protocol to assess exercise time. This protocol includes the use of an incremental workload that has the disadvantage that small differences in exercise capacity cannot be detected.
Medical treatment, mortality, and morbidity
After TMLR medical treatment did not change substantially with the exception of decreased sublingual nitroglycerin use which was not observed in the control group. One patient died postoperatively because of a myocardial infarction (perioperative and 1-year mortality of 6.7% vs 0% in the control group). Our perioperative mortality was slightly higher than the perioperative mortality reported in the other five randomized trials (1%–5%). Excluding cross-over patients, our 1-year mortality after TMLR was comparable with that in the other five randomized trials (5%–21%). Including cross-over in these five studies the overall 1-year mortality after TMLR was 74 out of 545 patients (13.6%) versus (excluding cross-over patients) 34 out of 355 patients (9.6%) receiving continued medication (from a total of 937 patients included in the five studies) [2–6].
In our study overall morbidity and hospital admittance was similar (and low) in both the TMLR and control group except for myocardial infarctions which occurred more often after TMLR (5 vs 0 infarctions in the control group). The morbidity (and mortality) after TMLR mainly occurred perioperatively. This emphasizes the importance of adequate selection and postoperative care in TMLR-treated patients. TMLR-induced ventricular arrhythmias were not observed. In our opinion the complication rate is acceptable given the impressive subjective improvement in the majority of TMLR-treated patients.
Myocardial perfusion and function
We did not find any changes in myocardial perfusion after TMLR. Both reversible and persistent defects remained unchanged. These results are in accordance with three [2, 3, 5] out of four [1–5, 7] previously reported randomized TMLR trials.
In contrast to myocardial perfusion scintigraphy, however, after TMLR stress echocardiography showed a small but considerable decrease in reversible wall motion abnormalities as well as an increase in fixed wall motion abnormalities. Our findings may reveal a reduced sensitivity regarding stress echocardiography for the detection of ischemia in this specific patient group. However this study was not designed to compare myocardial perfusion scintigraphy and stress echocardiography head-to-head for the detection of myocardial ischemia. Moreover stress echocardiography was analyzed according to perfusion territories whereas myocardial perfusion scintigraphy used a segmental model. Therefore explanations for the noted differences remain speculative and probably reflect differences related to the imaged substrate (ie, perfusion alone vs perfusion-related function).
The observed small increase in fixed wall motion abnormalities after TMLR did not correlate with perioperative myocardial infarction or myocardial infarction during follow-up. In addition the summed rest score on the myocardial perfusion scintigraphy remained unchanged at follow-up. This increase in fixed wall motion abnormalities may alternatively reflect the presence of dysfunctional but viable myocardium. However the stress echocardiography was performed without a low-dose dobutamine protocol and without special attention to the detection of viable myocardium. Therefore the explanation that the small increase in fixed wall motion abnormalities could be due to small differences in stunned or hibernating myocardium remains hypothetical.
The overall minimal change in global myocardial function after TMLR and in the control group seems clinically irrelevant because mean values at follow-up stay within normal limits. This minimal change most likely reflects the progression of disease.
Further comparison with other lasers
TMLR conducted using a CO2, holmium:YAG, or excimer laser results in laser-specific formation of channels with a particular shape and diameter as well as with varying degrees of myocardial thermal and mechanical damage [10]. Animal studies performed by us and by others have demonstrated that the use of these three lasers provide different results in terms of myocardial damage and TMLR-induced changes in vascular density and in myocardial function with possibly less favorable results after excimer TMLR in comparison with CO2 and holmium:YAG TMLR [17, 18]. Nevertheless whatever the working mechanism of TMLR may be, differences in laser-tissue interaction and tissue response after TMLR do not seem to translate into important differences in therapeutic efficacy. Overall in four of the earlier randomized studies that used a CO2 or holmium:YAG laser, TMLR was recommended [1, 3–7]. In one study, because of a (not significantly) higher mortality after TMLR, TMLR was not advocated [2].
In general our results are in accordance with those of other studies, as in all randomized trials [1–7], including ours, TMLR was effective in relieving angina and in most trials an improved QOL was demonstrated. For many clinicians this is sufficient reason to use this modality. Our results using an excimer laser might be viewed in a similar manner. Given the small differences in efficacy and the lack of a randomized comparison of the various lasers, the costs, the quality of technical support, and the viability of the supplying company are currently factors that predominantly determine the choice of laser for TMLR.
Patients with a high perioperative risk
This and other randomized studies have demonstrated that with adequate patient selection TMLR can be performed safely with a mortality and morbidity that is comparable with CABG. However in patients with a high perioperative risk due to severe comorbidity or a very low LVEF, we believe that other experimental approaches such as spinal cord stimulation [19] should be considered initially. Randomized studies are required to establish whether such therapies are similarly effective while maintaining a reduced risk.
Conclusion
In conclusion treatment of refractory angina with excimer TMLR provides results similar to earlier CO2 and holmium:YAG studies showing a relief of angina and improved QOL without evidence of improved cardiac perfusion or function. Consequently TMLR is primarily a symptomatic treatment with results that are comparable with other approaches including revascularization strategies. Therefore in patients who exhibit severe refractory angina, TMLR is a valuable option.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Katrapati and J. C. George Vineberg Operation: A Review of the Birth and Impact of This Surgical Technique Ann. Thorac. Surg., November 1, 2008; 86(5): 1713 - 1716. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
= deceased.