Quality of life and survival after transmyocardial laser revascularization with the holmium:YAG laser

CARDIOVASCULAR

Quality of Life and Survival After Transmyocardial Laser Revascularization With the Holmium:YAG Laser Kristine J. Guleserian, MD, Hersh S. Maniar, MD, Cindy J. Camillo, RN, Marci S. Bailey, RN, Ralph J. Damiano, Jr, MD, and Marc R. Moon, MD Division of Cardiothoracic Surgery, Washington University School of Medicine, St. Louis, Missouri

Background. The purpose of the this investigation was to assess postoperative survival and quality of life with transmyocardial laser revascularization (TMR) in highrisk patients. Methods. During a 24-month period, 81 consecutive patients underwent either sole therapy TMR (n ⴝ 34) or TMR with coronary artery bypass grafting (n ⴝ 47) using a holmium:yttrium–aluminum– garnet (YAG) laser. Outcomes were assessed in three high-risk groups, including patients with left ventricular dysfunction (ejection fraction < 0.40) (n ⴝ 37), unstable angina (n ⴝ 30), and congestive heart failure (n ⴝ 33). Disease-specific quality of life was assessed using the Seattle Angina Questionnaire in 58 late survivors and compared with an agematched cohort undergoing coronary artery bypass grafting only (no TMR) (n ⴝ 20). Results. Overall mortality was 6% ⴞ 3% (ⴞ70% confidence limit) and appeared higher with left ventricular dysfunction (11% ⴞ 5% vs 2% ⴞ 2%), but the difference

did not reach statistical significance (p ⴝ 0.17; power ⴝ 0.16). There was also no statistical difference with unstable angina (10% ⴞ 6% vs 4% ⴞ 3%; p > 0.53) or congestive failure (9% ⴞ 5% vs 4% ⴞ 3%; p > 0.66). However, survival at 18 months was significantly lower with left ventricular dysfunction (62% ⴞ 9% vs 90% ⴞ 5%; p < 0.003) and congestive failure (48% ⴞ 10% vs 96% ⴞ 3%; p < 0.001). For sole therapy TMR, quality of life was diminished comparing TMR with coronary artery bypass grafting (p < 0.004) and coronary artery bypass grafting only (p < 0.002). Conclusions. Transmyocardial laser revascularization can be performed in high-risk patients, but survival is significantly impaired in patients with left ventricular dysfunction and congestive failure, and quality of life is diminished without some degree of direct revascularization. (Ann Thorac Surg 2003;75:1842– 8) © 2003 by The Society of Thoracic Surgeons

S

aluminum– garnet (YAG) laser as either sole therapy (34 patients) or in conjunction with coronary artery bypass grafting (CABG) (47 patients) at Washington University Medical Center (Barnes-Jewish Hospital). There were 46 men (57%) and 35 women (43%), with a mean age (⫾1 standard deviation) of 61 ⫾ 11 years (range, 34 to 85 years). Three-vessel disease was present in 46 patients (57%) and 11 (14%) had left main disease. Forty-two patients (52%) had undergone previous CABG, including 97% of sole therapy TMR and 19% of TMR with CABG (p ⬍ 0.002). Only 1 patient in this series underwent primary TMR without prior or concomitant CABG. Diabetes mellitus (59%), hypertension (88%), hypercholesterolemia (79%), smoking history (64%), and prior myocardial infarction (70%) were present in the majority of patients. High-risk patient groups were defined as those with LV ejection fraction (EF) (ⱕ 0.40), unstable angina, or CHF. Mean LV EF was 0.44 ⫾ 0.14 and was less than or equal to 0.40 in 37 patients (46%). Thirty-three patients (41%) had CHF. Thirty patients (37%) had unstable angina, 13% of whom (4 patients) were taken to the operating room for urgent revascularization. All patients were contacted for follow-up by mail or telephone during a 2-month closing interval ending May 2002. Cumulative long-term follow-up totaled 1,150 patient-months and was 100% complete.

everal prospective randomized studies have demonstrated that transmyocardial laser revascularization (TMR) can decrease angina symptoms, improve exercise tolerance, and diminish the rate of rehospitalization when compared with maximal medical therapy in patients with end-stage coronary artery disease who are not suitable candidates for either percutaneous or open surgical revascularization [1– 4]. However, patients with left ventricular (LV) dysfunction, unstable angina, and congestive heart failure (CHF) have been excluded from many studies based on the assumption that their initial operative risk may have been prohibitive. The purpose of this investigation was to assess operative mortality and early postoperative (18-month) survival in high-risk versus low-risk patients undergoing TMR, and to determine the impact of TMR on postoperative quality of life (QOL).

Patients and Methods From February 2000 to March 2002, 81 consecutive patients underwent TMR using the holmium:yttriumPresented at the Forty-ninth Annual Meeting of the Southern Thoracic Surgical Association, Miami, FL, Nov 7–9, 2002. Address reprint requests to Dr Moon, Division of Cardiothoracic Surgery, Queeny Tower, Suite 3108, One Barnes-Jewish Hospital Plaza, St. Louis, Missouri 63110-1013; e-mail: [email protected].

© 2003 by The Society of Thoracic Surgeons Published by Elsevier Inc

0003-4975/03/$30.00 PII S0003-4975(03)00178-4

Quality of Life Sixty-two patients were alive at the time of late follow-up, of which 58 (94%) completed the appropriate surveys and were included in the late QOL study group. Two patients were excluded because of dementia (a 72 year-old man) and incarceration (a 44 year-old man), and 2 patients were excluded because of limited follow-up (⬍ 3 months). Of the QOL study group, sole therapy TMR was performed in 24 patients and TMR with CABG in 34 patients. Data were also collected from an age-matched, postoperative time-matched cohort of patients who had undergone CABG only (no TMR; n ⫽ 20). The three late follow-up groups were similar in age (p ⬎ 0.45), gender distribution (p ⬎ 0.88), preoperative CHF (p ⬎ 0.69), unstable angina (p ⬎ 0.81), LV EF (ⱕ 0.40) (p ⬎ 0.89), diabetes mellitus (p ⬎ 0.11), three-vessel disease (p ⬎ 0.66), and left main disease (p ⬎ 0.67). As expected, the incidence of previous CABG operation was significantly higher with sole therapy TMR (96%) than TMR with CABG (24%) or with CABG alone (15%) (p ⬍ 0.001). In addition, prior myocardial infarction was more common in sole therapy TMR (75%) than with only CABG (35%) (p ⬍ 0.02) but not TMR with CABG (56%) (p ⬎ 0.22). To compare the three late follow-up groups, diseasespecific QOL was assessed using the Seattle Angina Questionnaire [5]. The questionnaire is composed of 19 items. Four domains of health were examined, including physical limitation resulting from coronary artery disease, frequency of angina symptoms, satisfaction with treatment, and the patient’s perception of how coronary disease limits their quality of life. The score for each domain ranges from 0 to 100, with higher scores indicating better function. A clinically important mean group change is between five and eight points [6]. Cumulative late follow-up totaled 420 patient-months in the sole therapy TMR group, 562 patient-months in the TMR with CABG group, and 407 patient-months in the CABG alone group. Mean follow-up was similar between the groups; 20 ⫾ 1 months for CABG alone, 18 ⫾ 6 for sole therapy TMR, and 17 ⫾ 6 for combined TMR with CABG (p ⬎ 0.13).

Operative Techniques Transmyocardial laser revascularization was performed with techniques similar to those described previously [7, 8]. Operative exposure was obtained through a left anterior thoracotomy for patients undergoing sole therapy TMR or through a median sternotomy for those undergoing combined TMR with CABG. Holmium:YAG laser (Eclipse Surgical Technologies, Sunnyvale, CA) was used with all patients delivering laser energy with a flexible 1-mm optical fiber at 7 watts and 5 pulses per second. For patients undergoing sole therapy TMR, revascularization channels were created every square centimeter throughout the distal two-thirds of the left ventricle. For those undergoing TMR with CABG, revascularization channels were created on the empty beating heart after institution of cardiopulmonary bypass, but before aortic crossclamp placement and coronary grafting. Revasculariza-

GULESERIAN ET AL QOL AND SURVIVAL WITH TMR

1843

tion channels were similarly placed every square centimeter in areas of myocardium determined to be unsuitable for bypass grafting based on the operative surgeon’s assessment of the preoperative angiogram and intraoperative anatomic findings. The average number of channels was 30 ⫾ 9 during sole therapy TMR compared with 17 ⫾ 6 during TMR with CABG (p ⬍ 0.001). Five patients in the TMR with CABG group also underwent a valve procedure, including mitral valve repair (4 patients) or aortic valve replacement (1 patient). Length of stay was 5.6 ⫾ 2.2 days for sole therapy TMR and 10.8 ⫾ 8.7 days for TMR with CABG (p ⬍ 0.005). The average number of bypasses in the TMR with CABG group was lower than in the CABG only group (2.3 ⫾ 0.9 vs 3.3 ⫾ 0.9; p ⬍ 0.003). Complete revascularization was accomplished when at least one graft was placed in each of the three major vascular regions distal to a 50% diameter narrowing, if one existed [9]. Revascularization was complete in 19 patients (95%) in the CABG alone group compared with only 15 patients (44%) in the TMR with CABG group (p ⬍ 0.01). The patients in the TMR with CABG group that underwent complete revascularization had adjuvant TMR to a smaller subregion that did not have a bypass vessel, but that was felt to be important with regard to the patient’s potential for angina relief.

Data Analysis Operative mortality included any death that occurred during the initial hospitalization or within 30 days of operation for discharged patients. Survival data included death from all causes. Continuous data are reported as mean ⫾ 1 standard deviation and were compared between groups using the Student’s t test or analysis of variance as appropriate. When indicated by a significant F statistic (p ⬍ 0.05), differences were isolated using Fischer’s protected least significant different test. Power analyses were performed with ␣ ⫽ 0.05. Clinically important ratios and odds-ratios are reported with 70% confidence intervals. Actuarial survival estimates were calculated using the Kaplan-Meier method and were compared using the log-rank test. Variability of the actuarial estimates is expressed as ⫾1 standard error of the mean. Univariate analysis (␹2 test) and multivariate stepwise regression analysis were used to determine the preoperative and intraoperative risk factors that were significant, independent predictors of survival, and the risk factors that influenced the four domains of the Seattle Angina Questionnaire (SigmaStat 2.03 [SPSS Inc, Chicago, IL]). Twenty-one variables were analyzed: age, gender, year of operation, hypertension, diabetes mellitus, hypercholesterolemia, pulmonary disease, cerebrovascular disease, peripheral vascular disease, chronic renal insufficiency, cigarette smoking, family history of coronary disease, prior myocardial infarction, anginatype (stable, unstable), congestive heart failure, LV EF (ⱕ 0.40), status (urgent, elective), extent of disease (single, double, triple), left main disease, prior CABG, and sole therapy TMR versus TMR with CABG.

CARDIOVASCULAR

Ann Thorac Surg 2003;75:1842– 8

1844

GULESERIAN ET AL QOL AND SURVIVAL WITH TMR

Ann Thorac Surg 2003;75:1842– 8

CARDIOVASCULAR Fig 1. Postoperative survival estimates for patients undergoing transmyocardial laser revascularization (TMR) as sole therapy (n ⫽ 34) or in conjunction with coronary artery bypass grafting (CABG) (n ⫽ 47). The numbers of patients at risk are indicated.

Results Operative Mortality The operative mortality rate was 6% ⫾ 3% overall (5 of 81 patients); 9% ⫾ 5% (3 of 34) for sole therapy TMR, and 4% ⫾ 3% (2 of 47) for TMR with CABG (p ⬎ 0.64; power ⫽ 0.06). Excluding the 5 patients who underwent concomitant valve repair or replacement decreased the mortality rate for the TMR with CABG group to 2% ⫾ 2% (1 of 42). The causes of operative death were ventricular arrhythmia (2 sole therapy TMR patients), myocardial infarction (sole therapy TMR), LV failure (TMR, CABG, mitral repair), and cerebrovascular accident (TMR with CABG). The operative mortality rate appeared higher for patients with LV EF (ⱕ 0.40) (11% ⫾ 5% vs 2% ⫾ 2% for patients with LV EF ⬎ 0.40), but the difference did not reach statistical significance (p ⫽ 0.17; power ⫽ 0.19). Similarly,

Fig 2. Postoperative survival estimates for patients with left ventricular (LV) ejection fraction (EF) greater than 0.40 (n ⫽ 44) and less than or equal to 0.40 (n ⫽ 37). The numbers of patients at risk are indicated.

Fig 3. Postoperative survival estimates for patients with congestive heart failure (CHF) (n ⫽ 33) and without (n ⫽ 48). The numbers of patients at risk are indicated.

there was not a significant difference in operative mortality for patients with unstable angina (10% ⫾ 6% vs 4% ⫾ 3%; p ⬎ 0.35; power ⫽ 0.09) or CHF (9% ⫾ 5% vs 4% ⫾ 3%; p ⬎ 0.39; power ⫽ 0.07).

Early Postoperative Survival Of the 76 operative survivors, there were 14 late deaths (8 cardiac, 6 noncardiac). Multivariate regression analysis identified two factors as independent predictors of late death: ejection fraction (ⱕ 0.40) (p ⬍ 0.02; odds ratio ⫽ 5.0; 70% confidence interval: 2.2 to 11.1) and CHF (p ⬍ 0.001; odds ratio ⫽ 14.7; 70% confidence interval: 6.8 to 31.7).

Fig 4. Seattle Angina Questionnaire (SAQ) scores (mean ⫾ 1 standard error of the mean) for patients who underwent sole therapy transmyocardial laser revascularization (TMR), combined TMR with coronary artery bypass grafting (CABG), or CABG alone an average of 18 ⫾ 6 months postoperatively (⫾ standard deviation). Diseasespecific domains included physical limitation, angina frequency, treatment satisfaction, and disease perception (p ⬍ 0.002 for all sole therapy TMR versus CABG only [analysis of variance] and p ⬍ 0.004 for all sole therapy TMR versus TMR with CABG).

Ann Thorac Surg 2003;75:1842– 8

GULESERIAN ET AL QOL AND SURVIVAL WITH TMR

1845

Number of patients Physical limitation Angina frequency Treatment satisfaction Disease perception

Sole Transmyocardial Laser Revascularization

Transmyocardial Laser Revascularization and Coronary Artery Bypass Grafting

Coronary Artery Bypass Grafting Only

24 39 ⫾ 25 53 ⫾ 34 75 ⫾ 26 48 ⫾ 27

34 71 ⫾ 26a 85 ⫾ 19a 91 ⫾ 13a 77 ⫾ 21a

20 80 ⫾ 22a 94 ⫾ 23a 95 ⫾ 13a 88 ⫾ 20a

a p ⬍ 0.003 versus sole therapy transmyocardial laser revascularization; p ⬎ 0.10 for all coronary artery bypass grafting only versus transmyocardial laser revascularization with coronary artery bypass grafting (Analysis of variance, Fischer’s protected least significant difference [PLSD] test).

Continuous variables are mean ⫾ standard deviation.

Actuarial survival rates (including operative deaths) for all patients were 88% ⫾ 4% at 6 months and 77% ⫾ 5% at 18 months and were not significantly different for sole therapy TMR versus TMR with CABG (p ⬎ 0.71; Fig 1). Survival was similar for unstable versus stable angina (18-month survival: 75% ⫾ 7% vs 80% ⫾ 8%; p ⬎ 0.72), but was significantly diminished with an EF (ⱕ 0.40) (18month survival: 62% ⫾ 9% vs 90% ⫾ 5%; p ⬍ 0.003; Fig 2) and CHF (18-month survival: 48% ⫾ 10% vs 96% ⫾ 3%; p ⬍ 0.001; Fig 3).

Quality of Life Quality of life was consistently lower in the sole therapy TMR group than in the other two groups across all four domains of the Seattle Angina Questionnaire (Fig 4, Table 1). In the sole therapy TMR group, physical limitation, angina frequency, treatment satisfaction, and disease perception were all diminished compared with CABG only (p ⬍ 0.001 for all) and TMR with CABG (p ⬍ 0.002 for all). When comparing the CABG only group with the TMR with CABG group, there was no significant difference in physical limitation (p ⬎ 0.26), angina frequency (p ⬎ 0.25), or treatment satisfaction (p ⬎ 0.38), but disease perception tended to be lower in the TMR with CABG group (p ⫽ 0.075 vs CABG only). Multivariate regression analysis identified independent predictors of diminished quality of life for each domain. The absence of CABG was a significant predictor of decreased QOL across all four domains (p ⬍ 0.002). Other significant risk factors included diabetes mellitus (physical limitation; p ⬍ 0.02), congestive heart failure (physical limitation; p ⬍ 0.003), poor LV function (angina frequency; p ⬍ 0.005; disease perception, p ⬍ 0.05), and increased age (disease perception, p ⬍ 0.04). When asked, in retrospect, if they would undergo the operative procedure again, 58% of patients responded “yes, definitely” in the sole therapy TMR group compared with 65% in the TMR with CABG group (p ⬎ 0.83) and 95% in the CABG only group (p ⬍ 0.02). Regarding the patients’ overall general perception of their current health status, in the sole therapy TMR group, 29% considered their health to be good or excellent, 33% fair, and 38% poor. This distribution was significantly worse than in the TMR with CABG group (50% good or excellent, 50% fair, 0% poor; p

⬍ 0.001) and CABG only group (80% good or excellent, 20% fair, 0% poor; p ⬍ 0.001) groups.

Comment Operative mortality rates reported after TMR vary widely from 0% to 20%, but in more recent larger series the mortality rates typically range from 3% to 12% [1– 4, 8, 10 –15]. Many studies, however, have excluded or limited patients with poor LV function, unstable angina, or CHF making extrapolation of the results to high-risk patient subgroups difficult. Frazier and colleagues[4] included patients with unstable angina in their randomized series and found that unstable angina was the sole predictor of perioperative mortality. When patients with unstable angina (8%) were excluded from their analysis, perioperative mortality fell from 3% to 1%. Frazier and colleagues [4] also noted that LV EF (ⱕ 0.45), which occurred in 52% of their patients, was a significant predictor of 12-month mortality, similar to the findings ofthe present study. In the current retrospective study, which included sole therapy TMR and TMR with CABG, overall mortality appeared to be higher in patients with LV EF (ⱕ 0.40) (11% ⫾ 5% vs 2% ⫾ 2%), but the difference did not reach statistical significance in univariate analysis (p ⬎ 0.17). Importantly, however, multivariate analysis identified LV dysfunction (p ⬍ 0.02) and CHF (p ⬍ 0.001) as predictors of late death within 2 years. Burkhoff and associates [13] reported an overall operative mortality rate of 12% ⫾ 3% with sole therapy TMR, but found no statistical difference between patients with unstable and stable angina (16% ⫾ 5% vs 9% ⫾ 4%; p ⬎ 0.32). Conversely, in a multicenter study, Hattler and associates [12] noted higher operative mortality with unstable angina (16% ⫾ 4% vs 3% ⫾ 2%; p ⫽ 0.005), but no difference in late mortality up to 1 year postoperatively among operative survivors (13% ⫾ 4% vs 11% ⫾ 4%; p ⫽ 0.83). A 1999 report from the Duke group [16] also suggested that unstable angina was associated with increased perioperative mortality, but only 6% of patients (2 of 34) presented with unstable angina. In the current report, the mortality rate with unstable angina was not significantly higher than with stable angina (10% ⫾ 6% vs 4% ⫾ 3%; p ⬎ 0.35) and 18-month survival was the same

CARDIOVASCULAR

Table 1. Results of Seattle Angina Questionnaire for Patients Undergoing Sole Therapy Transmyocardial Laser Revascularization, Combined with Coronary Artery Bypass Grafting, or Standard Coronary Artery Bypass Grafting Only

1846

GULESERIAN ET AL QOL AND SURVIVAL WITH TMR

CARDIOVASCULAR

(75% ⫾ 7% vs 80% ⫾ 8%; p ⬎ 0.72). Furthermore, when looking exclusively at patients undergoing sole therapy TMR, the similarity in mortality rates with unstable (n ⫽ 12) and stable (n ⫽ 22) angina remained consistent (8% ⫾ 8% vs 9% ⫾ 6%; p ⬎ 0.57). Interestingly, Tjomsland and coauthors [14] identified diabetes mellitus as a risk factor for morbidity and mortality after sole therapy TMR. In addition, Krabatsch and colleagues [15] reported less angina relief in diabetic patients, but they found no survival difference compared with patients without diabetes mellitus. In the current series, diabetes mellitus did not impact operative survival significantly (8% ⫾ 4% with vs 3% ⫾ 3% without; p ⬎ 0.61) although the power of the statistical analysis was not high.

Quality of Life In the current study, postoperative QOL scores were significantly lower in the sole therapy TMR group than in the TMR with CABG group. Given that TMR with CABG patients received fewer transmyocardial channels compared with sole therapy TMR, and given that the preoperative characteristics were similar between the groups (except for the incidence of prior CABG), it can be inferred that the discrepancy in late QOL was the result of direct myocardial revascularization, although it was incomplete in 56% of patients in the TMR with CABG group. Interestingly, QOL was not significantly different between the CABG only group and TMR with CABG group in the physical limitation, angina frequency, or treatment satisfaction domains. Although the comparisons between these two groups were presented as a case-controlled analysis, the results are consistent with those of Allen and coauthors’ [8] prospective randomized study, which failed to show any difference in symptomatic relief 12 months after CABG either with or without TMR. The current series was different in that patients in the CABG only group consistently underwent complete revascularization (95%), whereas only 44% of those in the TMR with CABG group underwent complete revascularization. In Allen and coauthors’ [8] multicenter trial, all patients underwent incomplete revascularization, and those in the CABG without TMR group were left with untreated ischemic regions. Clearly it is important that some degree of direct myocardial revascularization be performed if possible (regardless of whether it is complete or incomplete) to maximize late quality of life.

Potential Limitations The holmium:YAG laser used in this study is routinely compared with the CO2 laser (PLC Medical Systems, Milford, MA), and substantial differences exist between these two technologies. Horvath and coauthors [17] and Spertus and coauthors [18] reported long-term success using the CO2 laser with respect to prolonged improvement in angina relief and QOL up to 5 years postoperatively. In contrast to the findings of De Carlo and coauthors [19] and Schneider and associates [20], Horvath and coauthors [17] used the holmium:YAG laser and found that 68% of patients experienced successful long-term angina relief at 5 years with 170% improvement in Seattle

Ann Thorac Surg 2003;75:1842– 8

Angina Questionnaire scores over base line. In a large series performed by a single investigator, Lansing [21], comparing patients undergoing TMR with CO2 (228 patients) and holmium:YAG (67 patients), angina relief at 1 year was more substantial with CO2 (40% no angina, 31% class I, 21% class II) than holmium:YAG (21% no angina, 27% class I, 45% class II). These studies suggest that there may be a difference in efficacy between various lasers that must be noted in studies examining the long-term results of TMR. In summary, the current report demonstrated that TMR, either as sole therapy or in conjunction with CABG, can be performed in high-risk patients, but the operative risk is not negligible and early postoperative survival is markedly diminished in patients with LV dysfunction and congestive failure. In addition, patients undergoing sole therapy TMR had significantly depressed QOL scores when compared with patients undergoing TMR with CABG or CABG only. Therefore, while some patients may experience excellent results after sole therapy TMR, it is important that some degree of direct myocardial revascularization be performed if possible to maximize late quality of life. The authors gratefully acknowledge the clinical contributions of Dr Thoralf M. Sundt III.

References 1. Schofield PM, Sharples LD, Caine N, et al. Transmyocardial laser revascularisation in patients with refractory angina: a randomised controlled trial. Lancet 1999;353:519 –24. 2. Allen KB, Dowling RD, Fudge TL, et al. Comparison of transmyocardial laser revascularization with medical therapy in patients with refractory angina. N Engl J Med 1999; 341:1029 –36. 3. Burkoff D, Schmidt S, Schulman P, et al. Transmyocardial laser revascularization compared with continued medical therapy for treatment of refractory angina pectoris: a prospective randomised trial. Atlantic Investigators Angina Treatments—Lasers and Normal Therapies in Comparison. Lancet 1999;354:885–90. 4. Frazier OH, March RJ, Horvath KA. Transmyocardial laser revascularization with a carbon dioxide laser in patients with end-stage coronary artery disease. N Engl J Med 1999;341: 1021–8. 5. Spertus JA, Winder JA, Dewhurst TA, et al. Development and evaluation of the Seattle angina questionnaire: a new functional status measure for coronary artery disease. J Am Coll Cardiol 1995;25:333–41. 6. Spertus JA, Winder JA, Dewhurst TA, Deyo RA, Fihn SD. Monitoring the quality of life in patients with coronary artery disease. Am J Cardiol 1994;74:1240 –4. 7. Sundt TM, Moon MR. Primary transmyocardial laser revascularization for end-stage coronary artery disease. Op Tech Thorac Cardiovasc Surg 2001;6:118 –31. 8. Allen KB, Dowling RD, DelRossi AJ, et al. Transmyocardial laser revascularization combined with coronary artery bypass grafting: a multicenter, blinded, prospective, randomized, controlled trial. J Thorac Cardiovasc Surg 2000;119: 540 –9. 9. Moon MR, Sundt TM, Pasque MK, Barner HB, Gay WA, Damiano RJ. Influence of internal mammary artery bypass grafting and completeness of revascularization on long-term outcome in octogenarians. Ann Thorac Surg 2001;72:2003–7. 10. Horvath KA, Mannting F, Cummings N, Shernan SK, Cohn

11.

12. 13. 14.

15.

16.

LH. Transmyocardial laser revascularization. Operative techniques and clinical results at two years. J Thorac Cardiovasc Surg 1996;111:1047–53. Milano A, Pratali S, Tartarini G, Mariotti R, DeCarlo M, Paterni G, et al. Early results of transmyocardial revascularization with a holmium laser. Ann Thorac Surg 1998;65: 700 –4. Hattler BG, Griffith BP, Zenati MA, et al. Transmyocardial laser revascularisation in the patient with unmanageable unstable angina. Ann Thorac Surg 1999;68:1203–9. Burkhoff D, Wesley MN, Resar JR, Lansing AM. Factors correlating with risk of mortality after transmyocardial revascularization. J Am Coll Cardiol 1999;341:55–61. Tjomsland O, Aaberge L, Almdahl SM, et al. Perioperative cardiac function and predictors for adverse events after transmyocardial laser treatment. Ann Thorac Surg 2000;69: 1098 –103. Krabatsch T, Petzina R, Hausmann H, Koster A, Hetzer R. Factors influencing results and outcome after transmyocardial laser revascularization. Ann Thorac Surg 2002;73:1888 – 92. Hughes GC, Landolfo KP, Lowe JE, Coleman RB, Donovan

GULESERIAN ET AL QOL AND SURVIVAL WITH TMR

17. 18.

19.

20.

21.

1847

CL. Perioperative morbidity and mortality after transmyocardial laser revascularization. Incidence and risk factors for adverse events. J Am Coll Cardiol 1999;33:1021–6. Horvath KA, Aranki SF, Cohn LH, et al. Sustained angina relief 5 years after transmyocardial laser revascularization with a CO2 laser. Circulation 2001;104(Suppl I):I-81–4. Spertus JA, Jones PG, Coen M, et al. Transmyocardial CO2 laser revascularization improves symptoms, function, and quality of life: 12-month results from a randomized controlled trial. Am J Med 2001;111:341–8. De Carlo M, Milano AD, Pratali S, Levantino M, Mariotti R, Bortolotti U. Symptomatic improvement after transmyocardial laser revascularization. How long does it last? Ann Thorac Surg 2000;70:1130 –3. Schneider J, Diegeler A, Krakor R, Walther T, Kluge R, Mohr FW. Transmyocardial laser revascularization with the holmium: YAG laser. Loss of symptomatic improvement after 2 years. Eur J Cardiothorac Surg 2001;19:164 –9. Lansing AM. Transmyocardial revascularization. Mechanism of action with carbon dioxide and holmium-yttriumaluminum-garnet lasers. J Thorac Cardiovasc Surg 1998;115: 1392.

DISCUSSION DR THORALF SUNDT (Rochester, MN): Congratulations for getting this work done while in the midst of a busy clinical residency. I know that you are not a research fellow, but are in fact carrying a heavy clinical responsibility. You have presented the data clearly and honestly. Having said that, I must, however, disagree with your conclusions. The data you have shown demonstrate an operative mortality rate in these “high risk patients” of 10%. That, for a therapy whose only clearly proven benefit is symptomatic relief of angina. Despite the fact that we have studied transmyocardial laser revascularization (TMR) clinically and in the laboratory for almost more than a decade, we still have no idea how it works. The only thing that is consistent from study to study is an improvement in angina. On every other point there continues to be debate. Improvement in perfusion is inconsistent from study to study; some say yes, some say no. The same can fairly be said of ventricular function and excercise tolerance. There is certainly no impact on long-term survival. Indeed you have made this last point with your own observations. In your series the long term survival of these “high risk patients” was dismal. So I think that you have presented the data fairly, but I think the words we use in making our conclusions are important. I would like to go on record as saying that I do not think that these data demonstrate an acceptable operative mortality for TMR in “high risk” patients. DR GULESERIAN: Thank you for your comments. DR GLENN PENNINGTON (Johnson City, TN): I enjoyed this presentation, but I share some of Dr. Sundt’s concerns. Were the procedures done for angina? Your abstract does not say they all had angina; it says some of them had unstable angina. DR GULESERIAN: They all had angina. DR PENNINGTON: So is that why the procedure was done? You were not doing this to treat heart failure, were you? DR GULESERIAN: No.

DR PENNINGTON: Okay. Of course, there are other lasers. Those who use the CO2 laser would claim that the holmium: YAG (holmium:yttrium–aluminum– garnet) laser causes significantly more damage to the tissue and, therefore, in a patient with severe heart failure it may not be optimal. Perhaps CO2 would be better. Can you comment on that at all? DR GULESERIAN: Thank you for your comment, and I agree, there have been several histologic studies that have shown that the laser channels created with the holmium:YAG laser are a lot larger. The channels created with the CO2 laser tend to be more discrete. Because our study did not really compare the two types of lasers, I can not really comment about the outcome in terms of the patients with congestive heart failure and what their mortality may have been. So I really cannot answer that question. I do want to go back a little bit to the operative mortality, and I think that in the conclusions when we do say that this procedure can be performed in patients with what we feel is acceptable mortality, I think that it is just acceptable mortality in comparison with other studies that have been published. Now, people have reported anywhere from 0% to 25% mortality with TMR, and although I agree with Dr Sundt that an operative mortality of 10% is not great, these are patients who really have no options, and it offers them some sort of therapy, and a lot of the studies looking at quality of life down the road, whether there may or may not be improved myocardial perfusion based on perfusion studies, oftentimes patients do “feel” like they are better. DR JAMES JONES (Columbia, MO): I enjoyed your article. I think you identify several very important points here. I have done 125 of these procedures as sole therapy, with one operative death. When we first studied the holmium:YAG, we were very concerned about operating on people who had unstable angina or recent cardiac events. As we have had more people in our intensive care units with recent problems, patients who could not be discharged and required a lot of resources, I have been increasingly willing to do these procedures earlier. In the last

CARDIOVASCULAR

Ann Thorac Surg 2003;73:1842– 8

1848

GULESERIAN ET AL QOL AND SURVIVAL WITH TMR

CARDIOVASCULAR

year I have done 10 within a couple of weeks of their cardiac events, and they have done well. I am not advocating that we operate routinely on unstable patients, but sometimes we must. The patient that I had die had an ejection fraction of about 30%. Otherwise, about 40% of my patients have had ejection fractions of less than 40%, and they seem to have done pretty well. The gentleman who died was the eighth patient in my series, and the autopsy suggested that I had made too many holmium:YAG laser channels. Also just mentioned, the holmium:YAG channel is 1 mm of vaporized tissue and then, at least in Burchoff’s dog studies, 2 mm of thermal acoustic injury. So if you add them all together you have about 5 mm of injured cardiac tissue, and if you leave less than a centimeter between channels, thinking more is better, you are going to have a piece of dead heart muscle. I think that this is probably the reason for many of the deaths and that this was the reason for my patient’s death. I would ask you then, how many laser channels did your group perform? Were there more channels made on patients

Ann Thorac Surg 2003;73:1842– 8

who did not survive? Also what was the identified cause of death among that group in your study? Thank you for your very nice study. DR GULESERIAN: Thank you for your comments. In the patients who underwent TMR as sole therapy, we had, on average, about 30 laser channels created in comparison with 17 in the group who underwent TMR coronary artery bypass grafting (CABG), and as you would expect, in areas that you are able to revascularize, you do not need as many laser channels. We did not look at the difference to see whether or not the number of channels made a difference in mortality. In our group, of the 5 deaths we had, 2 were from ventricular arrhythmias, 1 was from a myocardial infarction, 1 was from a left ventricular failure (that was in a patient who had a concomitant mitral valve repair), and 1 was from a stroke, which was also in a patient who underwent TMR CABG, but the two ventricular arrhytmias and the myocardial infarction were in the group who underwent TMR as sole therapy.