Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Is transmyocardial revascularisation of benefit to people with 'no option' angina?

Three Part Question

In [patients with 'no option' angina] is the use of [transmyocardial revascularisation] of any benefit in terms of [angina relief or survival]?

Clinical Scenario

You are seeing a 67 year old diabetic with CCS grade IV angina. A perfusion scan demonstrated clear areas of reversible ischaemia. Unfortunately you have carefully reviewed the angiogram with a colleague and although there is significant disease in all three regions, the disease is diffuse and there are no graftable vessels. You cannot offer this patient a coronary arterial bypass graft which disappoints the patient greatly, as he has no quality of life currently. A colleague has recently been to America and came back reporting that the use of TMR was widespread. Your colleague has therefore recently acquired a Holmium:YAG laser. You wonder whether to offer this option to your patient, but you resolve to look up the reported benefit and the mortality risk first so the patient can be adequately informed.

Search Strategy

Medline 1966-June 2004 using the OVID interface.
[ OR OR OR OR exp laser surgery/] AND [ OR exp angina pectoris, variant/ OR exp angina pectoris/ OR exp angina unstable/] LIMIT to human studies.

Search Outcome

A total of 345 abstracts were found from Medline of which 11 were relevant. These are summarized in the table.

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Burns et al,
International transmyocardial laser revascularization registry, containing 932 procedures from 15 centres. 78% of patients were angina class III or IV 90% had ejection fraction of >30% 177 procedures were combined TMR and CABG 24 were described as emergency procedures mean no channels 28Multicentre cohort study (level 2b )Mortality/Morbidity90/932 patients died (9.7%) 29% of patients had a complication, 8% bleeding, 4% MI, 8% LVFBreak down of results in CABG+TMR group not performed. Only 30% of patients in the registry has pre-op and post-op angina scores and the exercise testing was performed in only 10% of the patients. 12% of patients required post-operative adrenaline, 8% required an IABP. Only 59% required no post-operative support.
6 minute exercise test improvement (reported by 4 centres)TMR groups improved their exercise test times by mean 1 min 50s at 12 months compared to pre-operatively. P<0.01
Improvement in Angina classCCS angina score improved by 2 or more classes in 34% of patients at 12 months, and NYHA angina score 2 classes in 49& at 12 months.
Burkhoff et al,
TMR versus medical management of 182 patients from 16 USA centres with CCS III-IV angina, with at least one area of protected myocardium and EF>30% TMR group N=92, MM N=90 Left thoracotomy incision, Holmium YAG laser Mean 18 channels (range 9-42) 1 year follow-upMulticenter PRCT (level 1b)Improvement CCS Angina Score after 1 year of 2 or more pointsTMR group 47/77 (61%) Medical management alone 8/73 (11%) P<0.001Angina scoring was conducted by independent investigators blind to Rx. They compared investigator assessment of angina with independent assessment and showed that investigators were biased in their reporting.
Exercise tolerance at 12 monthsTMR group 65 sec median increase MM group 46 sec drop P=0.0001
Mortality at 1 yearTMR 5/92 deaths (5%) Medical management 9/90 (10%) P<0.001
Allen et al,
212 patients from 9 USA centres with refractory CCS IV, EF>25%, and evidence of reversible ischaemia. TMR=100, MM = 112 Holmium YAG Laser 40 +/- 8 channels 5 year follow-upMulticentre PRCT (level 1b)Improvement in CCS Angina Score after 5 year of 2 or more points, according to blinded assessmentTMR alone 42/48 (88%) MM alone 16/36 (44%) P<0.001 Mean angina class in TMR group at 5 years was 1.2 +/- 1.177% of the 275 patients originally entered into the study were successfully followed up to 5 years. 90% of patients had received previous CABG or PCI. 26% of medically managed patients became 'unstable' and received TMR during the study, and the operative mortality for this was 7%. 40% of medically managed patients had a post-enrolment procedure, vs 22% in TMR group
5 year intention to treat survivalTMR alone 33/50 (65%) MM alone 26/49 (52%) p=0.03

Operative mortality 3%
Bridges et al,
5 RCTS on TMR alone and 1 RCT On TMR+CABG Found after searching Medline and contacting expertsSystematic review (level 1a)Recommendations for TMR

Classifications I-Conditions with general agreement of effectiveness of treatment
II-Conditions with conflicting evidence of effectiveness of treatment
IIA-Weight of evidence is in favour of usefulness
IIB-Efficacy is less well established by evidence
III-Conditions with general agreement that treatment is harmful/not useful

Level of evidence
A-Data from multiple randomised control trials
B-Data from single randomised control trials or several non randomised studies
C-Consensus expert opinion
Recommendations for TMR as sole therapy
Class I patients with EF>30 and CCS III/IV refractory to medical therapy (level A)
Class IIB patients with EF<30 with or without IABP (level C)
Patients with unstable angina requiring intravenous medication (level B)
Patients with CCS II angina (level C)
Class III patients with acute MI, cardiogenic shock, VT/SVT, decompensated CHF (level C)
Keith Allen and Keith Horvath, 2 major advocates of TMR were on this committee, which may have biased the recommendations in favour of TMR
Allen et al,
275 patients at 18 USA centres with medically refractory CCS IV angina, demonstrable reversible ischaemia on perfusion scanning, EF>25%, and stable clinical status were eligible TMR N=132, MM N=143 Holmium YAG Laser 39 +/- 11 channels Limited left anterior thoracotomy. 3-5 channels placed and digital pressure applied for 2 mins, then repeat across distal two-thirds of LVMulticenter PRCT (level 1b)Improvement in CCS Angina Score after 1 year of 2 or more points. Assessment was conducted by an independent laboratoryTMR 58/76 (76%) MM alone 16/50 (32%) P<0.00186% of patients had received previous CABG, prior to entry to the study. 32% of patients treated medically became unstable, and could not be weaned from iv anti-anginals over a 48 hr period. They were assigned as treatment failures and were then given TMR. The TMR procedure in these patients had a 9% mortality No difference in survival, or myocardial perfusion demonstrated by this study. (1 year survival 84% TMR, 89% medical therapy)
Treatment failure (defined as death, Q wave MI, 3 hospitalisations over 12 months failure to wean from iv anti-anginals over a 48 hr period)TMR 43% MM 73% P<0.001
Peri-operative complicationsTMR 7/132(5%) deaths, 6 non-Q wave MIs 16 episodes of ventricular arrhythmis
Schofield et al,
188 patients from a single UK centre with refractory angina CCS III and IV and EF>30%, were randomised to receive TMR and Medical management or Medical management (MM) alone TMR N= 94, Medical managemt N=94 Small antero-lateral thoracotomy 1000W CO2 Laser 30 channels 1-year follow upSingle centre PRCT (level 1b)Improvement in CCS Angina Score after 1 year of 2 or more pointsTMR 18/74 (25%) MM alone 3/78 (4%) p<0.001Loss to follow up of 15% by 1 year . 90-95% of patients had previous CABG in the two groups. Angina assessment was performed by non-independent study physicians. Only 27% of patients had grade IV angina
1 year survivalTMR and MM 89% MM alone 96% (p=0.14)
preoperative mortality 5% (5/94)
Exercise toleranceTMR treadmill exercise time 40s longer (95% CI -15 to 94secs) (p-0.152)
TMR 12min walk distance 33m further ((-7 to 44) (p=0.108)
Complications33% of TMR had wound or respiratory infections. 15% LVF
Hattler et al,
167 patients at 13 USA centres with unmanageable unstable angina (UUAG) and chronic angina (CA) randomised to receive to receive TMR. Unstable angina defined as patients admitted to CCU and unable to be weaned from iv nitrates for 7 days with at least 3 failed attempts to wean nitrates. UUAG+TMR=76, CA + TMR =91 Left Thoracotomy 1000W CO2 laser 25 +/- 10 channels 1-year follow upMulticenter cohort study (level 2b)Improvement in CCS Angina Score after 1 year of 2 or more pointsUUAG and TMR 9/37(24%)
CA and TMR 5/15 (33%)
90% of patients had previous CABG, 38% of unstable patients had EF<45%. Assessment of angina was not independent. 11% wound infection rate in UUAg group and 2 patients returned for bleeding in each group.
Perioperative mortalityUUAG and TMR 12/76 (16%)
CA and TMR 3/91 (5%)
Mortality at 1 year (days 31-365 days )UUAG and TMR 8/64 (13%)
CA and TMR 10/88 (11%) P=0.83
Aaberge et al,
100 patients in a single Norwegian centre with NYHA functional class III or IV and EF>30% Left thoracotomy TMR N =50, MM N= 50 800W CO2 Laser 48 +/- 7 channels 3-5 year follow-upSingle centre PRCT (level 2b)Improvement in NYHA Angina Score after 5 year s of 1 or more points using a standardised telephone interviewTMR 23/38 (61%)
MM alone 9/37 (24%)
There was a higher number of hospitalisations due to heart failure and a higher use of diuretics and ACE-I in the TMR group. Non-independent data collection
Improvement in NYHA Angina Score after 5 year s of 2 or more points using a standardised telephone interviewTMR 24%
MM alone 3%
MortalityTMR 8/38 (22%)
MM alone 9/37 (24%)
4% operative mortality
Frazier et al,
192 patients from 12 USA centres with CCSIII and CCIV with EF>20% angina were randomised to receive TMR or Medical management (MM) TMR=91, MM =101 850W CO2 Laser 36 +/- 13 channels 1 year follow-upMulti-centre PRCT (level 1b)Improvement in CCS Angina Score after 1 year of 2 or more pointsTMR 72%
MM alone 13%
MR but crossed over to TMR 43%
Angina assessment was performed independently 60 medical patients crossed over to TMR therapy after 'treatment failure' Freedom from unstable angina or MI was higher in the TMR group
MortalityTMR 13/91 (12%)
MM 22/101 (22%) of which 15 deaths were in the crossover group
3 intra-operative deaths
ComplicationsTMR patients 7% MI, 11% CCF, 8% VT or VF
Peterson et al,
1998-2001 Database from the Society of Thoracic Surgeons with TMR procedures from 173 US centres, were compared with those from six published randomised control studies RCTdatabase: TMR N=722 TMR+CABG N=263 STS database: TMR N=661 TMR+CABG N=2,475Meta analysis (level 2a)Mortality for TMR aloneTMR RCT group 25/722 (3.5%)
TMR STS group 42/661 (6.4%)
Data limited to the STS database, which is only two-thirds of all US centres 36% of all US sites that contribute to the database now perform a median of 12 TMR procedures (range 1-150) annually, mostly combined with CABG. No post-operative data recording angina status was available from the STS database.
Mortality for TMR+CABGTMR+CABG RCT group 4/263 (1.5%)
TMR+CABG STS group 104/2475 (4.2%)
Mortality for all CABG alone vs all CABG+TMRCABG alone 1602/39064 (4.9%)
CABG and TMR 19/390 (4.1%)
Horvath KA,
200 patients from 8 hospitals, 80% patients had grade IV angina, 82% previous CABG. Perfusion scans performed on all patients Left anterolateral thoracotomy thorugh 5th ics. Co2 laser average 25+/-9 holes made.Cohort study (level 3b)Reduction in angina of 2 classes or more117/156 (75%) at 3 months, 70/95 (75%) at 12 monthsOnly 95 patients followed up to 12 months. No control group Angina scoring not independently assessed A significant decrease in the number of perfusion defects on perfusion scanning was found.
Morbidity2% MI, 1 patients had Mitral valve damage requiring repair, 1% bleeding rate, 4% IABP rate.
Perioperative mortality18/200 (9%) died


Bridges et al in 2004 issued a guideline for the Society of Thoracic Surgeons for TMR, stating that there was general agreement that any patients with CCS grade III or IV angina with maximal medical therapy not amenable to revascularisation, should undergo TMR as long as the ejection fraction was more than 30%. This was a very pro-TMR document with 2 members of the guideline committee being major advocates and active researchers in TMR. They reviewed 5 RCTs and the USA retrospective cohort, however they did not consider the negative results from Burns et al from the European Registry. Burns et al in 1999 reported the results of an international registry which was set up in 1994 to collect data on this emerging procedure. They reported 9377 procedures from 15 centres and found an overall 10% mortality, 30% complication rate and only 34% of patients had an improvement of CCS class of more than 2 classes at 12 months. Of the randomized controlled trials, the Atlantic trial was conducted in 16 US centres in 1999. 182 patients were randomized to TMR by a left lateral thoracotomy or medical management alone. These patients all had demonstrable reversible ischaemic regions, but also at least one area of protected myocardium (mostly from previous revascularisation) and an ejection fraction of over 30%. Five patients died in the TMR group, and only 1 within 30 days of the procedure. Nine died within one year in the medical management group. Independent assessors were used to measure angina at 2 years. CCS score was II or lower in 48% of patients but only 14% in the medical management alone group. It is interesting to note that the non-independent assessors (ie the surgeons) graded 32% of TMR patients as having a lower angina score than the independent assessors but only 11% higher. This systematic bias disappeared for the medical management group. This is an important finding when reading the other studies as this shows a clear problem with assessment of patients post-operatively by non-independent, operating surgeons. The Atlantic trial was a well conducted study that showed good results with a low mortality for TMR. Allen et al in 2004 published their 5 year follow up of their trial of TMR versus medical therapy in 275 class IV angina patients 8. They found that the mean angina score at 5 years was 1.2 in the TMR group, and that 88% of patients had a 2 class or more improvement in angina compared to 44% in the medically managed group. Allen also demonstrated a survival benefit, with a 65% 5 year survival in the TMR group versus a 52% survival in medically managed patients. However only 99 of the original 275 patients were followed up to 5 years, and 40% of medically managed patients ended up having a surgical procedure, including TMR for 26% of the patients. Although there were many confounders in this study, their results are still impressive for TMR. Of note this study extended the findings from the original study published in the New England Journal of Medicine in 1999, reporting the 1 year findings. The most significant negative randomized controlled trial came from Papworth in 1999. They studied 188 patients randomized to TMR via a small anterolateral thoracotomy or medical management alone. The primary outcome measure was the maximum exercise time and they found no improvement in this or the maximum walking distance. They found that only 25% of patients had an improvement of CCS angina score of more than 2 classes, and together with the 5% mortality they concluded that TMR cannot be advocated and should not be performed outside the context of a trial. This was a well conducted trial published in the Lancet but the angina scores were not independently assessed, and other authors have criticised the fact that only 27% of patients in this study had grade IV angina. Aaberge et al reported the 5 year follow up findings of 100 patients randomized to TMR or conservative treatment in a Norwegian PRCT. They had a 4% operative mortality, but found that 24% had at least a 2 point improvement in CCS angina score at 5 years compared to a 3% improvement in conservatively managed patients. They also found that the rate of heart failure treatment had increased in the TMR group and there was no effect on Ejection 10 fraction or mortality. Frazier performed a multicentre PRCT in 192 patients from 12 US centres. They found that with only a 3% perioperative mortality, angina improved by 2 or more grades in 72% of patients at 1 year compared to only 13% of patients in the medical management group. They also found some benefits in perfusion defects but no changes in survival. It should be noted that this study was complicated by the fact that 60 of the 101 medically managed patients had 'failed treatment' and underwent TMR. Hattler et al reported the findings of a multicentre cohort study comparing TMR for chronic angina with TMR for unstable angina in patients who were taken from the CCU after 7 days of iv nitrates and 3 failed attempts to wean this therapy. They found that in the 76 patients who had unstable angina, there was a 16% perioperative mortality compared to 5% in the chronic angina group, and only a 25% improvement of more than 2 CCS grades, although angina was not assessed independently. They reported that TMR was still a possible option in these otherwise 'no option' patients, although this mortality rate does seem very high and there was no control group to determine what the conservative management outcome would have been. Peterson et al in 2003 reported the results of the Society of Thoracic Surgeons database, and compared this with the published RCTs on this topic. Interestingly they found that TMR is being increasingly performed in the USA and now 36% of US centres now perform TMR, although this is mostly in combination with CABG. They found that the operative mortality of TMR alone in the STS database was 6.4% compared to the reported mortality of only 3.5% in the RCTs. They also reported that mortality was significantly higher in patients with unstable angina, depressed ventricular function and patients with a recent MI. The Society of Thoracic Surgeons in the USA recommend TMR in selected 'no option' patients. This recommendation is supported by all but one of the RCTs that addressed this subject, where significant improvements in angina were consistently found, with a preoperative mortality risk of around 5%.

Clinical Bottom Line

In selected stable patients with 'no option' CCS grade III-IV angina TMR can significantly reduce the grade of angina at the cost of a peri-operative mortality of around 5%.


  1. Burns SM, Sharples LD, Tait S, et al. The transmyocardial laser revascularization international registry report. Eur Heart J 1999;20:31-37.
  2. Burkhoff D, Schmidt S, Schulman SP, et al. Transmyocardial laser revascularisation compared with continued medical therapy for treatment of refractory angina pectoris: a prospective randomised trial. ATLANTIC Investigators. Angina Treatments-L Lancet 1999;354:885-890.
  3. Allen KB, Dowling RD, Angell WW, et al. Transmyocardial revascularization: 5-year follow-up of a prospective, randomized multicenter trial. Ann Thorac Surg 2004;77:1228-34.
  4. Bridges CR, Horvath KA, Nugent WC, et al. Society of Thoracic Surgeons practice guideline series: transmyocardial laser revascularization Ann Thorac Surg 2004;77:1494-1502.
  5. Allen KB, Dowling RD, Fudge TL, et al. Comparison of transmyocardial revascularization with medical therapy in patients with refractory angina. N Eng J Med 1999;341:1029-1036.
  6. Schofield PM, Sharples LD, Caine N, et al. Transmyocardial laser revascularisation in patients with refractory angina: a randomised controlled trial. Lancet 1999;353:519-524.
  7. Hattler BG, Griffith BP, Zenati MA, et al. Transmyocardial Laser Revascularization18 in the patient with unmanageable angina. Ann Thorac Surg 1999;68:1203-1209.
  8. Aaberge L, Rootwelt K, Blomhoff S, et al. Continued symptomatic improvement three to five years after transmyocardial revascularization with CO(2) laser: a late clinical follow-up of the Norwegian Randomized trial with transmyocardial revascu J Am Coll Cardiol 2002;39:1588-1593.
  9. Frazier OH, March RJ, Horvath KA. Transmyocardial revascularization with a carbon dioxide laser in patients with end-stage coronary artery disease. N Eng J Med 1999;341:1021-1028.
  10. Peterson ED, Kaul P, Kaczmarek RG, et al. 19From Controlled Trials to Clinical Practise: Monitoring Transmyocardial Revascularization Use and Outcomes. J Am Coll Cardiol 2003;42:1611-1616.
  11. Horvath KA, Cohn LH, Cooley DA, et al. Transmyocardial laser revascularization: results of a multicenter trial with transmyocardial laser revascularization used as sole therapy for end-stage coronary artery disease. J Thorac Cardiovasc Surg 1997;113:645-653.