Long-Term Outcomes After CABG With Concomitant CO2 Transmyocardial Revascularization in Comparison With CABG Alone

Eldaif, Shady M. MD; Lattouf, Omar M. MD, PhD; Kilgo, Patrick MS; Guyton, Robert A. MD; Puskas, John D. MD; Thourani, Vinod H. MD

Innovations: Technology & Techniques in Cardiothoracic & Vascular Surgery:
doi: 10.1097/IMI.0b013e3181d85935
Original Article

Objectives: Transmyocardial revascularization (TMR) has been used as an isolated or adjunctive revascularization therapy in patients presumed to have nonbypassable coronary artery disease. The purpose of this study is to evaluate the short- and midterm mortality for patients with complete revascularization using TMR and coronary artery bypass grafting (CABG) compared with those patients with incomplete CABG revascularization and to document long-term follow-up in patients receiving TMR + CABG.

Methods: Seventy TMR + CABG patients were cohort matched with 70 patients undergoing isolated CABG with circumflex coronary artery disease, but with no bypassable distal targets, from 1999 to 2005 at Emory University Hospital. The data were retrospectively reviewed from a database after being prospectively entered. Results are presented in mean ± standard deviation, and Kaplan-Meier curves were created for long-term all-cause mortality.

Results: The TMR + CABG patients had a similar incidences to the CABG only group for preoperative ejection fraction (50.9 ± 11.2% vs. 50.7 ± 10.3%, P = 0.93), number of grafts (2.6 ± 1.1 vs. 2.5 ± 1.3, P = 0.5), and number of diseased vessels (2.8 ± 0.3 vs. 2.9 ± 0.4, P = 0.26). Off-pump surgery was used more often in the CABG alone group versus the TMR combined with CABG group (74.3% vs. 41.4%, P < 0.001). Postoperatively, there was no statistical difference among the TMR + CABG and the CABG alone groups for intensive care unit length of stay (4.3 ± 7.8 days vs. 2.6 ± 3.4 days, P = 0.026), postsurgical length of stay (7.6 ± 6.1 days vs. 6.8 ± 4.5 days, P = 0.31), stroke events (1.4% vs. 1.4%, P = 1.00), myocardial infarction (4.3% vs. 2.9%, P = 0.65), and 30-day mortality (5.7% vs. 4.3%, P = 0.70). Long-term survival rate was not statistically significant. In addition, 4-year follow-up in the TMR + CABG group had symptom improvement with reduction in New York Heart Association classification for class III/IV (P < 0.0001, baseline vs. 4-year follow-up).

Conclusions: The combination of TMR and CABG for complete revascularization is safe and carries no further risk to patients compared with CABG only. CABG + TMR patients tend to have increased resource utilization. Long-term follow-up shows similar survival between the groups. TMR can be a useful adjunct to CABG for complete revascularization.

Author Information

From the Division of Cardiothoracic Surgery, Clinical Research Unit, Emory University School of Medicine, Atlanta, GA USA.

Accepted for publication February 5, 2010.

Supported by an unrestricted research grant by Edwards Lifesciences, Corp, Irvine, CA USA to Vinod H. Thourani, MD.

Presented at the Annual Scientific Meeting of the International Society for Minimally Invasive Cardiothoracic Surgery, Boston, MA USA, June 11-14, 2008.

Address correspondence and reprint requests to Vinod H. Thourani, MD, Division of Cardiothoracic Surgery, Emory University School of Medicine, Emory University Hospital Midtown, 6th floor Medical Office Tower, 550 Peachtree Street NE, Atlanta, GA 30308 USA. E-mail: vthoura@emory.edu.

Article Outline

Because of the considerable improvement of medical therapies in the treatment of ischemic heart disease, patients are deferred to coronary artery bypass grafting (CABG) at later stages in the disease process and, at times, after other interventions.1 An increasing percentage of these patients cannot be completely revascularized by percutaneous coronary intervention (PCI) or CABG. As incomplete revascularization is an important predictor of adverse perioperative events and postoperative mortality,2,3 transmyocardial revascularization (TMR) has emerged as an important alternative for patients not amenable to conventional PCI or CABG.4,5 Other studies supported the findings that TMR therapy alone provides these patients superior angina relief, decreased hospitalizations, better quality of life, and event-free survival relative to medical therapy.6,7

TMR involves creating transmyocardial channels from the epicardial to the endocardial surface of the ischemic heart with multiple pulses of energy. The channels close shortly after their creation with minimal scarring.8 Although the mechanism of action has not been clearly defined, angiogenesis is commonly promoted as an explanation for the increased perfusion and better contractility of TMR-treated hearts.9–12

Considering the success of TMR sole therapy, its combined use has been investigated, with numerous groups reporting on the efficacy and safety of TMR as an adjunct to CABG.5,13–17 In fact, the combined use of TMR to CABG is included in the Society of Thoracic Surgeons (STS) practice guidelines.18 However, there are few reports showing single institution outcomes of adjunctive TMR relative to CABG and, in particular, even fewer reports documenting long-term follow-up >3 years.15,19 Therefore, we compared short-term major adverse cardiac events and long-term mortality between CABG patients with completed revascularization with adjunctive TMR and a case-matched incompletely revascularized cohort receiving CABG alone. Moreover, we report long-term follow-up on the combined CABG and TMR patients viewing reasons for hospitalization, symptomology, and New York Heart Association (NYHA) class.

Back to Top | Article Outline


Design, Setting, and Patient Selection

This is a cohort-matched study of outcomes among patients receiving CABG and TMR (hereafter CABG + TMR) compared with patients receiving CABG alone. In compliance with the Health Insurance Portability and Accountability Act regulations and after Institutional Review Board approval granted by Emory University, the institutional adult cardiac database, where data are prospectively entered, was retrospectively queried for all patients who underwent CABG + TMR and a matched primary CABG with circumflex coronary artery disease (CAD) but with no bypassable distal targets (CABG alone) at Emory University Hospital from 1999 to 2005, which yielded 70 patients in each group. Patient follow-up was obtained with telephone interviews. Extracted records from this retrospective, single-center cohort study included demographic data, preexisting comorbidities, surgeon identity, operative strategy, and clinical outcomes.

Back to Top | Article Outline
Surgical Technique

Each patient underwent a single surgical session consisting of either CABG + TMR or CABG alone. Transmyocardial laser channels were done using the carbon dioxide (CO2) TMR Heart Laser (a CO2 TMR laser manufactured by PLC Systems, Inc., Franklin, MA, and distributed by Edwards Lifesciences, Irvine, CA) and were placed with echo guidance. Conventional on-pump CABG was performed with standard techniques, using membrane oxygenators, roller head pumps, cardiotomy suction, arterial filters, cold antegrade and retrograde blood cardioplegia, and moderate hypothermia. Off-pump CABG was performed with one of several commercially available cardiac positioning and coronary artery stabilizing devices, using techniques previously described.20

Back to Top | Article Outline
Long-Term Follow-up

The Social Security Death Index is a national database of death records extracted from the United States Social Security Administration's Death Master File Extract. Persons with a Social Security Number who have died since 1963 and whose death has been reported to the Social Security Administration will be listed in the Social Security Death Index. Thus, for each patient who died before the cutoff date of March 31, 2007, a mortality date was provided, allowing construction of Kaplan-Meier (KM) long-term survival curves. This study contains 1 month worth of inhospital data (April 2007). Cause of death is not available; thus, this study seeks to describe all-cause long-term mortality in those patients undergoing CABG alone.

Back to Top | Article Outline
Variables of Interest

To reduce selection bias, the 70 CABG + TMR patients were matched with 70 controls (from a patient pool consisting of 121 CABG alone patients) using the greedy matching algorithm. This algorithm matches the 70 cases to controls using 70 consecutive decisions based on the shortest multivariate distance between the matching factors. Once a match is made, the pairing is never reconsidered. The matching factors in this study were patient age, ejection fraction, diabetes, peripheral vascular disease, and renal failure, and each received an equal weighting. Outcomes of interest were operative death, permanent stroke, myocardial infarction, renal insufficiency, intensive care unit (ICU) length of stay (LOS), and LOS from surgery to discharge. There were no missing data points among the matching factors.

All data for patients were entered into a surgical database, using the data definitions of the STS National Adult Cardiac Database. The quality of the data was checked both at the institutional level and before final entry into the STS national adult cardiac database.

Back to Top | Article Outline
Statistical Analyses

Patients were classified according to whether they received CABG + TMR or CABG alone. The χ2 tests were used to determine whether rates were different across groups. Group comparisons of continuous variables of interest were performed primarily with two-sample Student t tests. Because of small samples, ICU LOS and LOS from surgery to discharge were compared among groups using a nonparametric Wilcoxon Rank-Sum Test. Because the groups were not significantly different with respect to the matching variables, no multivariable adjustment was performed.

Long-term survival comparisons between treatment groups were made using KM product-limit estimates. KM curves were generated, which provide survival estimates at postoperative points in time. A log-rank test was used to determine whether long-term survival differed between the treatment groups. In addition, a Cox proportional odds regression model was constructed to identify predictors of long-term survival using the preoperative variables of interest (see Table 1). Hazard ratios (HRs) and 95% confidence intervals (CIs) for significant model terms were reported.

To determine whether there had been a significant shift in NYHA classification from baseline to year 4 follow-up, a McNemar's test for paired proportion was used. The data were analyzed using SAS version 9.1 (Cary, NC). All tests were two sided using an α = 0.05 level of significance. No adjustments for multiple tests were made.

Back to Top | Article Outline


Preoperative characteristics for both groups are listed in Table 1. The CABG + TMR patients had incidences similar to the CABG alone group for all preoperative characteristics including preoperative ejection fraction (50.9 ± 11.2% vs. 50.7 ± 10.3%, P = 0.96), number of grafts (2.7 ± 1.1 vs. 2.5 ± 1.3, P = 0.41), and number of diseased vessels (2.9 ± 0.4 vs. 2.8 ± 0.4, P = 0.37).

Similarly, the groups were well matched with respect to operative characteristics (Table 2). Off-pump surgery was used more often in the CABG alone group versus the CABG + TMR (74.3% vs. 41.4%, P < 0.001). In the CABG + TMR group, the number of channels ranged from 14 to 23, (mean = 17.3) and depended on the extent of myocardium that could not be bypassed, which was mainly in the posterolateral wall.

Back to Top | Article Outline
Inhospital Outcomes

Postoperatively, there was a statistical difference among the CABG + TMR and the CABG alone groups for ICU hours (103.0 ± 187.4 days vs. 58.8 ± 67.7 days, P = 0.026). Postsurgical LOS (7.6 ± 6.1 days vs. 6.8 ± 4.5 days, P = 0.22), stroke events (1.4% vs. 1.4%, P = 1.00), myocardial infarction (4.3% vs. 2.9%, P = 0.65), and 30-day mortality (5.7% vs. 4.3%, P = 0.70) were all statistically similar among the groups.

Back to Top | Article Outline
Four-Year Clinical Outcomes for TMR and CABG Group

For the CABG + TMR group, the mean follow-up time was 32.8 months. Four patients (5.7%) sustained inhospital deaths. Two CABG + TMR patients (2.8%) died within 6 months from the time of surgery and six others (8.6%) died between 18 and 29 months of follow-up. As for readmissions, a total of 21 readmissions (30%) were reported over the total follow-up period. Six patients (8.6%) were readmitted within 30 days for reasons including gastritis and gastrointestinal bleed, hypovolemia from overdiuresis, electrolyte abnormality, lower extremity wound infection, and suicide attempt. Only three readmissions (4.3%) were cardiac related: two patients presented with chest pain and one patient had transient arrhythmia. At the end of the follow-up period, 43 of the 59 living patients were assessed with phone interviews documenting their current medications and use of home oxygen, and NYHA classifications were assigned based on answering specific questions regarding their level of activity. Of the 43 patients, 25 patients (58%) reported mild shortness of breath (SOB) without the need for O2 use. At baseline, 19 patients (44%) were classified as NYHA class III/IV, of whom 11 were reclassified to class I/II, while 8 were class III/IV at 4 years follow-up (see Fig. 1). This shift was statistically significant (P < 0.001).

Back to Top | Article Outline
Long-Term Survival

Unadjusted long-term survival was statistically similar among the groups using the log-rank test (P = 0.19). Survival times by groups are listed in Table 3. In the Cox proportional hazards regression model, only renal failure was a predictor of long-term survival (HR = 6.54, 95% CI = 2.24-18.87, P < 0.001). After multivariable adjustment, the two groups are statistically similar with respect to long-term survival (HR = 1.41, 95% CI = 0.28-1.81, P = 0.47; Fig. 2; Table 4).

Back to Top | Article Outline


Currently, surgical CABG and PCIs are the procedural options most commonly available to treat patients suffering from ischemic heart disease.21 Revascularization in these patients results in symptomatic and survival benefit.2,3 However, many subjects with advanced CAD remain symptomatic, because their ischemic and viable areas of the myocardium are underperfused, accentuating the resulting left ventricular dysfunction.1 For patients with coronary anatomy not suitable for such procedural intervention, TMR has become an accepted standard of care.4–8 TMR is an often-used therapy for treatment of refractory angina in patients with ungraftable vessels.4–8 With TMR, a laser apparatus creates transmyocardial channels in a viable ischemic area of myocardium. The channels eventually heal shut, producing remarkably little scarring and encouraging new capillaries to grow in the area.10–12,22

The combination of TMR and CABG has been evaluated recently and included in the STS practice guidelines under class II, and recent consensus statements demonstrate adjunctive TMR + CABG can be recommended to improve long-term angina relief to reduce 30-day mortality and major adverse cardiac event and to improve 1-year exercise performance in patients with diffuse CAD who cannot be completely revascularized by CABG alone.18,23,24 These guidelines are based on reports documenting both safety and efficacy of this combined treatment. More specifically, several studies prospectively reviewed outcomes for the combined treatment versus CABG alone with up to 1-year follow-up.14,17,26,27 The published studies tended to be multicentered in nature where the centers are known for and attract referrals of TMR therapy. One aim of our study was to evaluate the success of this combined treatment in a single-institution setting where results can be more generalized.

Even fewer reports discuss the increased risk compared with CABG alone group with incomplete revascularization and show over 3-year long-term follow-ups.19,26 In this study, patients in the TMR adjunct group had slightly higher renal disease and had a trend toward having higher rates of hypertension compared with the CABG alone group. Otherwise, the groups were similar in their preoperative characteristics. In a study looking at preoperative characteristics predicting operative mortality and events, unstable angina, which was shown to increase mortality rates in sole-therapy TMR, was shown not to be a strong predictor of adverse events.16 For postoperative results, there were no significant differences between our two groups except for the TMR + CABG group tended to have longer ICU stay but, overall, similar timing for hospital discharge. Another study showed that there are increased mortality rates in the CABG alone group (7.6% vs. 1.5% in the combined TMR group), and the authors found that CABG alone and increased age were associated with higher mortality rates when investigating multivariate predictors of operative mortality.14 The combined procedure also showed survival advantage in another prospective study, which reported overall higher mortality rates in their groups (33% in the CABG alone group vs. 9% in the TMR adjuvant therapy group) because of dealing with sicker patients in their selection process.15,26 Our other short-term results were comparable with that published in other data series; however, there was no survival advantage in the TMR + CABG group observed in our study as was seen in the above-mentioned reports.

Long-term follow-up with the combined treatment was reported by Allen et al19 using a 20-W pulsatile Holmium laser TMR and showing no survival benefit at 5 years. Our analysis of long-term survival showed similar results among the groups as well, with no survival benefit to the combined group. Multivariable regression analysis showed that renal failure and age were predictors of long-term mortality. These two parameters, along with ejection fraction and diabetes, were predictive of long-term mortality in Allen et al's study.19 As for the long-term benefit of the combination, follow-up for the patients in the TMR adjuvant group showed a decrease in symptoms and improved NYHA classification, although this was not compared with the control group or could this benefit be linked to either CABG or TMR; however, we believe that complete revascularization contributes to this long-term additive benefit.

This study has several limitations. Its retrospective nature and limited sample size do not permit complete accounting for all sources of bias, specifically in determining the matched controls despite prospective data entry. Our matching method described previously still had a significant difference among patients revascularized using on-pump and off-pump. We did not show the differences in the angina scores between the groups unlike other similar studies; our intention was to show that the mortality and complication difference between these groups because follow-up for CABG alone patients is clearly established in the literature. However, we wanted to present our long-term data on the combined therapy. Our method of assessing function and NYHA classification was dependent on relying on patients' answers in our phone interviews and not a more objective test such as exercise tolerance. Finally, we did not consistently obtain heart perfusion scans; thus, this was not mentioned in this report; however, a previous study done in our institution reported increased perfusion to the myocardium with sole TMR and supported earlier work done by Frazier et al5 showing this therapeutic benefit for TMR (in Lattouf O, Sigman S. A study to assess improvement in myocardial perfusion after transmyocardial revascularization. Presented at the 16th World Congress of the World Society of Cardiothoracic Surgeons, Ottawa, Canada; 2006).

In conclusion, complete revascularization can be accomplished in the ischemic myocardium of CABG patients by using adjuvant TMR to otherwise ungraftable poor target areas. The combination of TMR and CABG carries no further risk to patients compared with CABG only. Short-term adverse events and long-term survival shows no further risk in adding TMR to the operative intervention. As the trend grows putting more emphasis on complete revascularization, TMR can be a useful adjunct to CABG to complete this task and potentially improve quality of life.

Back to Top | Article Outline


The authors thank the members of the cardiothoracic clinical research unit for their help with database searches and patient long-term follow-up.

Back to Top | Article Outline


1.Jessup M, Brozena S. Heart failure. N Engl J Med. 2003;348:2007–2018.
2.Osswald BR, Blackstone EH, Tochtermann U, et al. Does the completeness of revascularization affect early survival after coronary artery bypass grafting in elderly patients? Eur J Cardiothorac Surg. 2001;20:120–125; discussion 125–126.
3.Weintraub WS, Jones EL, Craver JM, Guyton RA. Frequency of repeat coronary bypass or coronary angioplasty after coronary artery bypass surgery using saphenous venous grafts. Am J Cardiol. 1994;73:103–112.
4.Allen KB, Dowling RD, Fudge TL, et al. Comparison of transmyocardial revascularization with medical therapy in patients with refractory angina. N Engl J Med. 1999;341:1029–1036.
5.Frazier OH, March RJ, Horvath KA. Transmyocardial revascularization with a carbon dioxide laser in patients with end-stage coronary artery disease. N Engl J Med. 1999;341:1021–1028.
6.Guleserian KJ, Maniar HS, Camillo CJ, et al. Quality of life and survival after transmyocardial laser revascularization with the holmium:YAG laser. Ann Thorac Surg. 2003;75:1842–1847; discussion 1847–1848.
7.Hattler BG, Griffith BP, Zenati MA, et al. Transmyocardial laser revascularization in the patient with unmanageable unstable angina. Ann Thorac Surg. 1999;68:1203–1209.
8.Horvath KA, Kadipasaoglu KA. Transmyocardial laser revascularisation. Lancet. 1999;353:1704–1705; author reply 1706–1707.
9.Domkowski PW, Biswas SS, Steenbergen C, Lowe JE. Histological evidence of angiogenesis 9 months after transmyocardial laser revascularization. Circulation. 2001;103:469–471.
10.Horvath KA, Chiu E, Maun DC, et al. Up-regulation of vascular endothelial growth factor mRNA and angiogenesis after transmyocardial laser revascularization. Ann Thorac Surg. 1999;68:825–829.
11.Horvath KA, Lu CY, Robert E, et al. Improvement of myocardial contractility in a porcine model of chronic ischemia using a combined transmyocardial revascularization and gene therapy approach. J Thorac Cardiovasc Surg. 2005;129:1071–1077.
12.Hughes GC, Lowe JE, Kypson AP, et al. Neovascularization after transmyocardial laser revascularization in a model of chronic ischemia. Ann Thorac Surg. 1998;66:2029–2036.
13.Actis Dato GM, Hakimpour M, Bacciega M, et al. TMR and CABG: the best way to obtain a complete and a more lasting revascularization? Ann Thorac Surg. 2000;69:1993–1995.
14.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 Thoracic Cardiovasc Surg. 2000;119:540–549.
15.Allen KB, Dowling RD, Richenbacher W. From controlled trials to clinical practice: monitoring transmyocardial revascularization use and outcomes. J Am Coll Cardiol. 2004;43:2364–2365; author reply 2365–2366.
16.Horvath KA, Ferguson TB Jr, Guyton RA, Edwards FH. Impact of unstable angina on outcomes of transmyocardial laser revascularization combined with coronary artery bypass grafting. Ann Thorac Surg. 2005;80:2082–2085.
17.Loubani M, Chin D, Leverment JN, Galinanes M. Mid-term results of combined transmyocardial laser revascularization and coronary artery bypass. Ann Thorac Surg. 2003;76:1163–1166.
18.Bridges CR. Guidelines for the clinical use of transmyocardial laser revascularization. Semin Thorac Cardiovasc Surg. 2006;18:68–73.
19.Allen KB, Dowling RD, Schuch DR, et al. Adjunctive transmyocardial revascularization: five-year follow-up of a prospective, randomized trial. Ann Thorac Surg. 2004;78:458–465; discussion 465.
20.Puskas JD, Williams WH, Duke PG, et al. Off-pump coronary artery bypass grafting provides complete revascularization with reduced myocardial injury, transfusion requirements, and length of stay: a prospective randomized comparison of two hundred unselected patients undergoing off-pump versus conventional coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2003;125:797–808.
21.Taggart DP, Kaul S, Boden WE, et al. Revascularization for unprotected left main stem coronary artery stenosis stenting or surgery. J Am Coll Cardiol. 2008;51:885–892.
22.Yamamoto N, Kohmoto T, Gu A, et al. Angiogenesis is enhanced in ischemic canine myocardium by transmyocardial laser revascularization. J Am Coll Cardiol. 1998;31:1426–1433.
23.Diegeler A, Cheng D, Allen K, et al. Transmyocardial laser revascularization: a consensus statement of the Internationals Society of Minimally Invasive Cardiothoracic Surgery (ISMICS) 2006. Innovations. 2006;1:314–322.
24.Cheng D, Diegleler A, Allen K, et al. Transmyocardial laser revascularization: a meta-analysis and systematic review of controlled trials. Innovations. 2006;1:295–313.
25.Deleted in text.
26.Frazier OH, Tuzun E, Eichstadt H, et al. Transmyocardial laser revascularization as an adjunct to coronary artery bypass grafting: a randomized, multicenter study with 4-year follow-up. Tex Heart Inst J. 2004;31:231–239.
27.Stamou SC, Boyce SW, Cooke RH, et al. One-year outcome after combined coronary artery bypass grafting and transmyocardial laser revascularization for refractory angina pectoris. Am J Cardiol. 2002;89:1365–1368.
Back to Top | Article Outline

This interesting report from the group at Emory University looked at 70 patients undergoing transmyocardial laser revascularization (TMR) and coronary artery bypass grafting (CABG) who were cohort controlled matched with 70 patients undergoing isolated CABG with circumflex disease and no bypassable distal targets. The combination of TMR and CABG did not add to morbidity or mortality. The long-term survival rate was not statistically significant between groups. This study did not address the issue as to whether TMR provided any broad benefit at late follow-up, but its addition to the surgery clearly did not appear to add risk. Readers are to be reminded that this study does have limitations in that it was retrospective in nature and, thus, subject to selection bias and involved a relatively small number of patients.


Transmyocardial; Revascularization; Coronary; Adjuvant

© 2010 Lippincott Williams & Wilkins, Inc.