This action might not be possible to undo. Are you sure you want to continue?
Effect of arterial revascularisation on survival: a systematic review of studies comparing bilateral and single internal mammary arteries
David P Taggart, Roberto D’Amico, Douglas G Altman
Background Coronary artery bypass grafting (CABG) is the commonest major operation in most developed countries. A single internal mammary artery (IMA) graft has proven survival benefits, but the additional survival advantage of a second graft is unknown. We systematically reviewed published studies of bilateral versus single IMA grafts in CABG to assess any differences in survival. Methods We identified from Medline all studies in which single and bilateral IMA grafts were compared. We included studies in which at least 100 patients in each group had been followed up for at least 4 years. We assessed study quality on the basis of patient selection, comparability of intervention groups (especially for age, sex, ventricular function, and diabetes status), outcome assessment, and completeness of follow-up. Our primary outcome was survival. Estimates of treatment effect (single versus bilateral) expressed as hazard ratios were pooled across studies. Findings None of the studies was a randomised trial, but nine cohort studies met our inclusion criteria. Seven studies yielded survival data for meta-analysis, and included 15 962 patients: 11 269 single and 4693 bilateral IMA grafts. The bilateral group had significantly better survival than the single group (hazard ratio for death 0·81; 95% CI 0·70–0·94). Exclusion of methodologically weak studies improved survival rates with bilateral IMA grafts. Interpretation Because no study was a randomised trial, our results are more uncertain than is indicated by the 95% CI. Nevertheless, bilateral IMA grafts seem to give better survival rates than single grafts. Lancet 2001; 358: 870–75
The clinical and prognostic benefits of coronary revascularisation for some subgroups of patients with ischaemic heart disease are well accepted,1 and worldwide every year around 1 million patients undergo coronary artery bypass grafting (CABG). CABG provides excellent short-term and intermediate-term outcomes, but long-term results are affected by vein graft failure. Most patients undergoing CABG need three bypass grafts; the standard operation uses a single internal mammary artery (IMA) and two vein grafts. Even without data from randomised trials, the clinical and survival advantages of a single IMA graft in CABG are widely accepted.1–3 By comparison with the use of vein conduits, a single IMA graft to the left anterior descending coronary artery improves survival and reduces the frequency of late myocardial infarction, recurrent angina, and need for further cardiac interventions.1–3 10 years after CABG, 90–95% of left IMA grafts are patent and disease free, whereas 75% of vein conduits are blocked or severely diseased.4 Veingraft failure leads to high risk of recurrent angina, late myocardial infarction, and need for further interventions;5 10–15 years after their first operation, 30% of patients might need CABG to be redone.6,7 Redo surgery is expensive and has more risks than first time surgery. Because vein grafts are more frequently used to graft non-left anterior descending coronary vessels, the inferior patency rates of such grafts might result from their use in coronary arteries with less favourable runoff. However, this explanation is refuted by several reports of long-term patency of single IMA anastomoses to nonleft anterior descending coronary vessels being far better than that obtained with vein grafts.8,9 Calafiore and colleagues8 reported angiographic patency rates of 99% in bilateral IMA grafts 18 months after CABG, with an estimated actuarial survival of 95% in 1800 patients at 8 years. Dion and colleagues9 reported that 96% of multiple arterial anastomoses to various coronary vessels were patent 7 years after CABG in 161 patients who consented to have repeat angiography. The possibility that bilateral IMA conduits result in clinical and survival benefits exceeding those of a single IMA graft has been examined by several groups.10–19 However, interpretation of these studies is complicated by: no randomisation; too few patients; and inadequate length of follow-up, completeness of follow-up, or both. Furthermore, because initial use of bilateral IMA was largely confined to young and low-risk patients, longterm benefits were attributed to the favourable survival characteristics of these patients, which obscured any potential benefit of bilateral IMA conduits per se. Because CABG is the most common major operation in the developed world, clinical and survival advantages, or
Oxford Heart Centre, John Radcliffe Hospital, Oxford, OX3 9DU, UK (D P Taggart PhD); ICRF/NHS Centre for Statistics in Medicine, Institute of Health Sciences, Oxford (R D’Amico PhD, D G Altman DSc); and Centre for Evaluation of Effectiveness of Health Care (CEVEAS), Modena, Italy, and Italian Cochrane Centre, Milan, Italy (R D’Amico PhD) Correspondence to: Mr David P Taggart (e-mail: email@example.com)
Copyright © 2001 All Rights Reserved
THE LANCET • Vol 358 • September 15, 2001
Two of us (RD and DA) assessed the quality of all included studies in accordance with six criteria in three domains: cohort selection. We decided to do a formal meta-analysis if there were studies of adequate methodological quality with sufficient outcome data. 16. with weights inversely proportional to the variance of the log hazard ratio of every report—ie. singly and in combination. Two of us (DT and RD) independently inspected about 3000 electronic reports identified by the searches. 13.16. preoperative data for age. or selection of group not described). †Inferred from length of study. diabetic status) were similar at baseline.12. Statistical analysis All outcomes considered were times to a specified event. 12=valvular reconstruction or replacement. ventricular function. We included data from only the last publication of centres that had produced sequential reports. The assessment scheme was based partly on the Ottawa-Newcastle system. DT and RD extracted all data independently and agreed on data inclusion after discussion. thoracic. We inspected the references of all studies to identify further studies. 14=non-coronary cardiac procedure preceded CABG. 3=operation for single-vessel disease. and diabetic status were sought. or arteries.11–13 we estimated the log hazard ratio and its variance from survival curves and tabulated results with Parmar and colleagues’21 method. 2001 Copyright © 2001 All Rights Reserved 871 . ‡Median. No star was assigned if the two groups differed. ventricular function. In three investigations.14. 11=patients undergoing emergency operation. How was the non-exposed cohort selected? (One star if drawn from the same community as the exposed cohort. 10=overseas patients. We estimated the hazard ratio with 95% CI for every study with adequate data. 13=large aneurysmectomy. multiple. We included published studies that had at least 100 patients in each group. In a further three. 5=synchronous or multiple endarterectomies. We defined survival as time from operation to death from any cause. sex. sex. and Number of patients SIMA Study (year) Johnson (1989)10 Morris (1990)11 Naunheim (1992)12 Dewar (1995)13 Berreklouw (1995)14 Pick (1997)15 Sergeant (1998)16 Buxton (1998)17 Lytle (1999)18 Farinas (1999)19 1438 420 100 765 143 161 4883 1557 8123 200 BIMA 576 643 100 377 143 160 1191 1269 2001 200 Length of follow up (mean years) SIMA ·· ·· 14 ·· 5‡ 9·8 ·· ·· 10·3 10 BIMA ·· ·· 14 ·· 5·1‡ 9·8 ·· ·· 10·1 8 Total >4† 4‡ ·· >4† ·· ·· >4 4·3 ·· ·· 1. Our primary outcome measure was survival. We obtained the overall log hazard ratio by computation of the weighted average of individual log hazard ratios. mammary.10 NR 15. 5. single. to establish whether use of bilateral IMA is beneficial. and had a median follow-up of at least 4 years. In particular. 6 12. sex. unilateral. both. in particular with reference to age. we sought preoperative data for age. Our primary outcome measure was survival. even if there were no other differences between the groups. 4. 17 NR NR 3. no star if more than 20% of patients were lost to follow-up. outcome (panel). of bilateral IMA would have substantial implications for individual patients and health-care systems. artery.ARTICLES Quality assessment of non-randomised studies Cohort selection was assessed on the answers to three questions Were details of criteria for assignment of patients to treatments provided? (We awarded one star for relevant details). Adequacy of cohort follow-up (one star if no patient or fewer than 20% of patients were lost to follow-up. 14 None 10. ventricular function. large studies with a smaller standard error were given more weight than smaller Exclusion criteria* Data available for meta analysis Number of deaths SIMA ·· 41 36 69 9 51 ·· 169§ 2107§ ·· BIMA ·· 76 19 44 5 30 ·· 98§ 407§ ·· Methods Procedures We searched Medline and Embase for publications containing the words internal. ventricular function. no star if drawn from a different source. *1=died within 90 days.17 estimation of the log hazard ratio and its variance were reported. §Estimated from summary information in report. or if the researchers did not provide relevant information). or multiple regression analysis was used to adjust for differences. How representative was the exposed cohort? (One star if representative of typical patient undergoing CABG. or diabetic status (two stars). bilateral. between 1972 (Medline) or 1980 (Embase) and 1999. 17=left anterior descending artery stenosis (<70%). Outcome was assessed by two criteria Assessment of outcome (one star for information ascertained by record linkage or interview. or differences controlled for. One star was assigned if one of these four characteristics was not reported. 16=left ventricular function (>15).20 in particular the use of stars awarded for each criterion. We included studies in which proper survival analysis methods had been used. cohort comparability. NR=not reported. sex. Table 1: Characteristics of studies THE LANCET • Vol 358 • September 15. 11 NR No Yes Yes Yes Yes Yes No Yes Yes No SIMA=single internal mammary artery. no star if groups of patients were selected or selection of group was not described). BIMA=bilateral internal mammary artery. and other characteristics had been controlled for. In particular. 6=reoperations. and diabetic status. Cohort comparability was assessed on the basis of study design or analysis of cohort differences No differences between the two groups. 4=alternate venous or arterial conduits were used. 15=age (>60) years. We aimed to identify studies reporting clinical outcome with single or bilateral IMA conduits. Meta-analysis was deemed possible a priori for those reports in which either the characteristics of patients (age. no star if this information was not reported or ascertained in some other way).
but groups were very similar with respect to other characteristics. In three reports. WMA=percentage of patients with moderate-to-severe wall motion abnormalities and EF<45%. Lytle and colleagues’ study18 was much larger than the others. mild. years) SIMA Study (year) Johnson (1989)10 Morris (1990)11 Naunheim (1992)12 Dewar (1995)13 Berreklouw (1995)14 Pick (1997)15 Sergeant (1998)16 Buxton (1998)17 Lytle (1999)18 ·· 60 51 62 55 62 ·· 65 58 BIMA ·· 60 50 61 54 60 ·· 59 55 Sex (% men) SIMA ·· ·· 87 83 88 80 ·· 78 84 BIMA ·· ·· 83 85 89 82 ·· 89 89 Ventricular function (mean or %) Classification ·· EF LVS WMA LVEDP EF ·· EF<50% No LVD Mild LVD Moderate LVD Severe LVD EF SIMA ·· 0·5 7·6 34·2 12·5 0·6 ·· 24% 34·5% 25·4% 15·8% 24·3% 0·6% BIMA ·· 0·5 7·6 30·4 13·5 0·6 ·· 5 36·5 27·6 13·8 22·1 0·57 Diabetes (%) SIMA ·· 16 3 19 9 27 ·· 20 13 BIMA ·· 21 4 18 7 17 ·· 7 11 Farinas (1999)19 56 53 89 84 10 17 SIMA=single internal mammary artery.11. BIMA=bilateral internal mammary artery. one surgeon practised an operative strategy of predominant single IMA grafting . LVEDP=mean left ventricular-end-diastolic pressure. None of these was a randomised trial.14. in theory.13. intention to treat—not ultimate therapy—was the basis of patient grouping”.19 Two of these reports10. . EF=mean ejection fraction.15 mean left ventricular score. Horizontal lines indicate 95% CI. Only three reports11. or severe). . because smaller studies have a larger standard error.16. retrospective. or presumed to be. and in one. . patient distribution to the care of each surgeon was based solely on random clinical consultation . Table 3: Assessment of quality of studies* 872 Copyright © 2001 All Rights Reserved THE LANCET • Vol 358 • September 15. moderate. Ten reports met our inclusion criteria (table 1). Morris and colleagues11 had not reported information about sex distribution. 2001 .13 number of patients lost during follow-up was not included (table 1). *Criteria for award of stars described in the panel. and diabetic status were similar in the bilateral and single IMA groups. as follows: “.13 mean left ventricular endStudy (year) Selection 1 Johnson (1989)10 Morris (1990)11 Naunheim (1992)12 Dewar (1995)13 Berreklouw (1995)14 Pick (1997)15 Sergeant (1998)16 Buxton (1998)17 Lytle (1999)18 Farinas (1999)19 ·· ✫ ·· ·· ✫ ·· ·· ·· ✫ ·· 2 ✫ ✫ ✫ ✫ ✫ ✫ ✫ ✫ ✫ ✫ 3 ✫ ✫ ✫ ✫ ✫ ✫ ✫ ✫ ✫ ✫ Comparability 4 ✫ ✫ ✫✫ ✫✫ ✫✫ ✫✫ ·· ✫✫ ✫✫ ✫✫ Outcome 5 ✫ ·· ✫ ·· ·· ✫ ✫ ✫ ✫ ✫ 6 ✫ ·· ✫ ·· ✫ ✫ ✫ ✫ ✫ ✫ diastolic pressure.19 the bilateral IMA group had significantly worse preoperative characteristics for which no quantitative adjustment was presented. LVD=left ventricular dysfunction. . .12 percentage with moderateto-severe wall-motion abnormalities and an ejection fraction of less than 45%. Table 2 shows the characteristics of patients in all ten studies.18 included information about process of treatment assignment. .13 used pair-matching to create groups of patients with similar characteristics. Three reports that met our basic criteria were excluded from the meta-analysis. and in two of these. Table 2: Characteristics of patients ones.14 and “The decision to use single or bilateral IMA grafts and the vessels that were chosen to be grafted with those IMAs were at the discretion of the attending surgeon”.18 In three reports. They did a detailed analysis with various Survival Hazard ratio Weight Favours Favours (95% CI) (%) BIMA SIMA Morris (1990)11 Naunheim (1992)12 Dewar (1995)13 Berreklouw (1995)14 Pick (1997)15 Buxton (1998)17 Lytle (1999)18 1·21 (0·84–1·73) 0·75 (0·45–1·26) 1·01 (0·58–1·72) 0·50 (0·18–1·40) 0·82 (0·50–1·33) 0·71 (0·56–0·91) 0·77 (0·66–0·89) 13·2 7·2 6·5 2·0 7·9 22·8 40·4 Overall (95% CI) 0· 2 0 5 0 5 1· 1· 2· 0· Hazard ratio 0·80 (0·70–0·94) Effects of bilateral IMA compared with single IMA Random-effects meta-analysis of data from seven studies. and the other surgeon attempted to maximise distal IMA anastomoses .14 and amount of left ventricular dysfunction (none.11 “Patients underwent bilateral ITA [internal thoracic artery] operation depending on preference of the surgeon”.11. ventricular function.10.12–14 mean age. and Cox’s model had been used in an attempt to adjust for this.18 Table 3 shows assessments of every report by the six criteria described in the panel. Results Our initial search of Medline produced over 3000 potentially relevant references.ARTICLES Age (mean.14 method of outcome assessment was not described. of which two12. None of the studies was randomised or included a description of unbiased choice of operation assignment. Information about ventricular function was reported in various ways that included mean ejection fraction. . sex distribution.16 did not include information about patient characteristics for each intervention group. thus it carried most weight in pooled analysis. . LVS=mean left ventricular score. thus we thought that an important imbalance in sex between the groups was unlikely.17–19 distributions of these variables differed between the two groups.15. and only one was clearly prospective. We used a random-effect meta-analysis.11 The other nine were identified as. In four.11.
In two other studies. Bilateral IMA grafting is generally used in several of the best European. only 9% of all eligible UK patients received two or more arterial grafts. For example. an analysis of intention-to-treat data in that respect. this apparent survival benefit should be interpreted cautiously because of the type of studies included in this systematic review. there is no obvious way to adjust for poor study design (or no design) or to quantify potential bias.14 gave a slightly stronger result in favour of bilateral IMA grafts (0·76.17 Discussion Our results show that bilateral IMA patients had a significant reduction in mortality compared with single IMA patients. are compounded if data are from non-randomised trials and are not reliable. Our main concerns were to understand how decisions were made for treatment assignment. we could not meta-analyse redo surgery rates—though five studies reported lower rates of reintervention with bilateral IMA grafts. although again. 0·67–0·86).22 Limitations of systematic reviews. but also on subtle factors that can be linked to prognosis.17 and American18 centres.17. even in relation to synthesis of results from randomised trials. and thus may strongly affect surgeons’ choice of graft. Some reassurance can be gained from studies with similar baseline characteristics in both groups. This type of grafting is technically more demanding than single grafts. and need for percutaneous angioplasty. and diabetic status. only Morris and colleagues11 indicated a system that was almost random. Poor application of the second conduit included use as a graft to the anatomically unimportant diagonal coronary artery. 2001 Copyright © 2001 All Rights Reserved 873 .18 Lytle and colleagues18 did the largest and best investigation. Comparison of interventions for which there are no randomised trials is often necessary. and the important elements of quality in non-randomised intervention studies have been researched. Omission of three methodologically weak investigations11. such adjustments might not fully remove confounding resulting from differences in prognostic characteristics. although not easily quantifiable. For similar reasons. such as recurrent angina.ARTICLES statistical methods. However. 95%CI 0·19–0·37). how comparable the two groups of patients were. Although differing presentation of data for redo surgery precluded formal meta-analysis. Results of all the studies combined indicated a significant reduction in mortality attributable to use of bilateral IMA (hazard ratio 0·81.15.15. Most studies did not include any indication of how treatment was chosen. However.10–19 and that such grafts have mortality of less THE LANCET • Vol 358 • September 15. we might. The factors that led surgeons to choose one operation instead of the other are not clear. ventricular function. in fact. hazard ratio 0·27. and better arterial graft patency than seen with vein grafts. There was no evidence of statistical difference between results of the seven studies. and because of the perception of an increased perioperative risk.18 No study results showed a significantly harmful effect of bilateral IMA grafts. age. Seven studies containing 15 962 patients (4693 bilateral and 11 269 single IMA patients) met our criteria for inclusion in a meta-analysis and had suitable data for mortality analysis.26 and its placement to the right coronary artery (which is nullified by subsequent development of disease at the crux). The figure shows the hazard ratios of bilateral and single IMA grafts.24 In essence. The surgeon chose the type of operation for every patient in the other studies (we included those that did not mention this issue).26 Longevity is improved by placement of both IMA grafts to the left side of the heart. have underestimated benefits of bilateral IMA grafts. because there were striking differences in patient characteristics in some reports but not in others. and showed that their results were robust to the different methods of analysis. However. We used adjusted results in studies without good compatibility at baseline. these workers compared one surgeon who favoured single IMA with another who preferred bilateral IMA—ie.27 Calafiore and colleagues8 reported 95% 8-year survival in more than 1800 patients with bilateral IMA grafts to the left side of the heart. which introduces bias from unmeasured variables.20. Our analysis might overestimate benefits of bilateral IMA grafts because of confounding by indication. and in three a lower incidence of reintervention was reported at 10 years. which adds further support to the hypothesis that leftside placement is best. but consistent evidence from previous studies indicates that bilateral grafts do not increase perioperative mortality or morbidity. Furthermore. in two studies a lower incidence of reoperation was reported in the bilateral IMA group14. in 1998. Deaths in each group for individual reports are shown in table 1.18 at 5 years after treatment. Sergeant and colleagues28 suggested that the reduction in incidence of redo surgery in patients who had received bilateral IMA conduits might be at least partly attributable to surgical reluctance to reoperate on patients with multiple patent arterial conduits because of increased technical difficulties. and how well the investigators had adjusted for baseline characteristics of patients in data analysis. We included unadjusted results in the meta-analysis only if in both groups important determinants of longterm survival were similar at baseline—ie. an increased reintervention rate was also reported in the single IMA group.13. but several intangible factors probably affect surgeons’ choice of operation.25. as a free graft from the aorta (which results in worse longterm patency rates). since the second IMA conduit was frequently used in what would now be thought a suboptimum fashion. 95%CI 0·70–0·94).17.9 Australian. The hazard ratio was lower than unity in five of the seven studies but was significant in only two. Insufficient data and different reporting techniques precluded meta-analysis of secondary outcome measures. but has not been widely adopted in the UK—eg.29 Bilateral IMA grafting is usually opposed because no randomised trials have been done. such comparability alone is clearly not a guarantee of similar prognosis. such confounding can occur if a specific treatment option is based not only on obvious clinical reasons.18. sex. and of the three that did.8.23. without data from randomised trials. the quality of coronary vessels requiring bypass grafting is an important determinant of graft survival. However. surgeons are not generally increasing their use of more than one arterial graft. Although a single IMA graft to the left anterior descending coronary artery has been established as the most important component of CABG.26. although the results seem precise because of the large total sample size and because studies had quite similar findings (and none showed any harmful effects of bilateral IMA). late myocardial infarction. at 10 years after treatment they reported a significantly lower rate of reoperation in bilateral than in single IMA patients (8% vs 40%.
Blackstone EH. 28: 616–26. Janusz MT. We have also lately reported that use of bilateral IMA grafts does not increase myocardial injury31 or lung injury32 in comparison with use of a single IMA graft. Eur J Cardiothorac Surg 2000. Dewar LR. Hooper GD. Leach HA. Jahari R. Unilateral versus bilateral internal mammary revascularization. Guyton RA. and follow-up of at least 10 years. Gordon I. The right internal thoracic artery graft-benefits of grafting the left coronary system and native vessels with a high grade stenosis. our 30–day mortality was 0·5% (Taggart D P. 64: 599–605. Circulation 2000. Ettiene PY. Long-term and angiographic follow up of sequential internal thoracic artery grafting. Does it make sense to use two internal thoracic arteries ? Ann Thorac Surg 1995. Meyns B. Dion R. Schaff HV. Ann Thorac Surg 1989. Contributors David P Taggart thought of the idea for this study. Weintraub WS. because without evidence from a randomised trial. Two internal thoracic artery grafts are better than one. 314: 1–6. Bilateral internal thoracic artery grafting may improve outcome of coronary artery surgery: risk-adjusted survival. Fiore AC. Jamieson WR. Circulation 1995. O’Connell D. et al. National adult cardiac surgical database report. Oxford. Brenowitz JB. Burton JR. J Thorac Cardiovasc Surg 2000. in particular Roberto D’Amico. Reid CM. Davis KB. Wells GA. 344: 563–70. Kayser KL. 2000. 2001. Carrier M. Stockman B. 18: 255–61. 92 (suppl 2): 8–13. Pick AW. Validation and interdependence with patient-variables of the influence of procedural variables on early and late survival after CABG. Paper presented at 3rd symposium on systematic reviews: beyond the basics. J Am Geriatr Soc 1999. David P Taggart wrote the initial draft of the report. Med Care 1995. et al.34 The lack of randomised controlled trials of surgical techniques35 makes assessment of best practice difficult. et al. J Am Coll Cardiol 1996. 14: 480–87. and a 1% significance level. Sergeant P. Schonberger JP. Loop FD. Sergeant P. Confirmation of even a small improvement in survival with bilateral IMA would be of major prognostic benefit for patients. Kinsman R. 120: 990–98. Eur J Cardiothorac Surg 2000. Rubay J. N Engl J Med 1996. challenges. 49: 202–09). 1998. Davidoff R. 73: 103–12. 120: 651–59. 82 (suppl 4): 214–23. Kafka HB. Farinas JM. 53: 716–18. Ann Thorac Surg 1992. Shea B. Tugwell P. Reoperative 25 26 2 27 3 28 4 29 5 30 6 31 7 coronary artery bypass surgery: early and late results and management in 1300 patients. Guyton RA. et al. et al. et al. Coronary bypass surgery with internal-thoracic-artery grafts: effects on survival over a 15–year period. Psaty BM. Measuring effects without randomized trials? Options. Green G. Derouck D. Miller CC. Ann Thorac Surg 1997. But such patients have the most to gain from bilateral IMA grafts. Meyns B. Peterson J. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Improved survival with multiple left-sided bilateral internal thoracic artery grafts. Glineur D. Eur J Cardiothorac Surg 1997. Circulation 1999. Cameron A. for The Society of Cardiothoracic Surgeons of Great Britain and Ireland. Fitzgibbon GM. 12: 1–19. Dion R. Downs SH. N Engl J Med 1986. 874 Copyright © 2001 All Rights Reserved THE LANCET • Vol 358 • September 15. All researchers helped design the study. 67: 466–70. The feasibility of creating a checklist for the assessment of methodological quality both of randomized and nonrandomized studies on health care interventions. Hennen B. Lytle BW. 1999. et al.388 patients during 25 years. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses. Comparison of long-term clinical results of double versus single internal mammary artery bypass grafting. Morris JJ. Demertzis S. Roberto D’Amico and David P Taggart extracted most of the data. Noirhomme P. Lytle BW. Influence of the internal mammary artery graft on 10–year survival and other cardiac events. there is a slight increase in risk of sternal wound problems in patients with diabetes in whom both IMAs are harvested. Biochemical assessment of myocardial injury after cardiac surgery: effects of a platelet activating factor antagonist. Bilateral internal thoracic artery grafting: long-term clinical and angiographic results of in situ versus Y grafts.33 and harvesting the IMA with a skeletonisation technique as opposed to a pedicle technique greatly reduces risk of wound problems. 98 (suppl): S1–S6. 33 (suppl): S8–S14. Fields BL. Eur J Cardiothorac Surg 1998. Craver JM. 47: 749–54. Calafiore AM. Barner HB. Loop FD. Eagle KA. et al. Keon WJ. Coronary bypass graft fate and patient outcome: angiographic follow-up of 5. Survival and eventfree performance. the technical requirements of multiple arterial grafting and the learning curve necessary to master such a complex technique are likely to prevent widespread uptake of bilateral grafting. Glower DD. Naunheim KS. Compete arterial revascularization in multivessel coronary artery disease with 2 conduits (skeletonized grafts and T grafts). Frequency of repeat coronary bypass or coronary angioplasty after coronary artery bypass surgery using saphenous vein grafts. Rosalion A. Ercan H. Ann Thorac Surg 1999. Ann Thorac Surg 1990. 52: 377–84. Tatoulis J. 4: 212–16. Circulation 1998. in our own institution. Orszulak TA. Anderson BJ. Fuller J. Peduzzi P. J Thorac Cardiovasc Surg 2000. Although we accept the inherent weakness of our data.18 Furthermore. Schmidt SE. 2001 . Effect of coronary artery bypass graft surgery on survival: overview of 10–year results from randomised trials by the Coronary Artery Bypass Graft Surgery Trialists Collaboration. Factors influencing longterm (10-year to 15-year) survival after a successful coronary artery bypass operation. Wyatt DA. et al. 36: 303–12. Stat Med 1998. Hebert Y. Buxton BF. Beal AC. All authors contributed to the final version. Cosgrove DM. ACC/AHA guidelines for coronary artery bypass graft surgery: executive summary and recommendations. Single versus bilateral internal mammary artery grafts: 10-year outcome analysis. Verhelst R. unpublished observations). a third arterial graft can also be used to achieve total arterial revascularisation. Assessment and control for confounding by indication in observational studies. 59: 1456–63. a combination of bilateral IMA and radial artery grafting) since June. analysed and compiled data. and with acceptably low morbidity. 117: 855–72. Vitolla G. and of economic benefit for health-care systems. in a consecutive series of over 200 patients undergoing total arterial revascularisation (ie. Blackstone E. Blackstone E. demonstration of a 5% reduction in mortality at 10 years with 90% power. Fuller JA. there is a suggestion that use of bilateral IMA conduits improves survival. et al. bilateral internal thoracic artery grafts. Schaff HV. 334: 216–19. Jones JW. 17: 2815–34. Taggart DP. Zucker D.065 grafts related to survival in reoperation in 1. Lin D. Black N. Cardiovasc Surg 1996. Wendler O. J Cardiovasc Surg 1995. Welch V. Am J Cardiol 1994. et al. Thornby JI. 17: 407–14 Johnson WD. et al. J Epidemiol Community Health 1998. First cardiological or cardiosurgical reintervention for ischemic heart disease and after primary coronary artery bypass grafting. 48: 19–25. and coronary endarterectomy. Circulation 1990. Parmar MKB. Kaul TK. J Thorac Cardiovasc Surg 1999. Jones EL. However. All researchers. 100: 1464–80. Berreklouw E. with a perioperative mortality rate of around 2%. 23 24 References 1 Yusuf S. Moses LE. Jones CR. El Khoury G. Update (1990: results of internal thoracic artery grafting over 15 years: single versus double grafts. problems. On the basis of our data.ARTICLES than 1% in appropriately selected patients. A report of the American College of Cardiology/American Heart Association task force on practice guidelines.30 Indeed. Komeda M. Kahn DR. Lancet 1994. Keogh BE. Torri V. Stewart L. Clinical evaluation of single versus multiple mammary artery bypass. Buxton BF. Berkshire: Dendrite Clinical Systems. Ruengsakulrach P. Smith R. 64: 9–15. Free internal mammary artery graft in myocardial revascularization. Such a trial would be timely. Contini M. 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Acknowledgments We thank Tom Treasure and Magdi Yacoub for reviewing this manuscript and for their constructive comments. 102 (suppl III): 79–83. would require randomisation of about 4200 patients. Koepsell TD. Ann Thorac Surg 1997.
Michael Winkler A 68-year-old female patient was admitted with fever and lassitude. J T Kielstein MD). Her blood pressure was 90/60 mm Hg. Subphrenic abscess is a classical cause of fever of unknown origin. Hannover.ARTICLES 32 Taggart DP. The psychiatric disease of our patient and her inability to give a history accounted for a difficult diagnosis in this case. Ann Thorac Surg 2000. Cultures grew Streptococcus viridans but echocardiographic studies failed to demonstrate any evidence of endocarditis. The patient made an uneventful recovery. Brooks MM. 35 Horton R. Lombardero MS. Department of Medicine (A Woywodt MD. pulse 120/min and temperature 39·2˚C. Clinical picture: Subphrenic abscess and rupture of the spleen Alexander Woywodt. Germany THE LANCET • Vol 358 • September 15. and Division of General Surgery and Transplantation. 33 Detre KM. Surgical research or comic opera: questions. She had an elevated leukocyte count and C-reactive protein. it transpired that the patient had received aspirin and low-molecular weight heparin and fallen repeatedly. She was alert but appeared confused and acutely ill. Examination of the specimen showed a large infected haematoma. but few answers. Jan Thomas Kielstein. Eur J Cardiothorac Surg 2000. The chest was clear but left subcostal tenderness was noted. and found a left-sided subphrenic collection of purulent fluid. Abdominal computed tomography (figure) showed a ruptured spleen (arrow) and free fluid below the left hemidiaphragm. Technical aspects of double-skeletonized internal mammary artery grafting. 342: 989–97. 34 Gurevitch J. Respiratory dysfunction after cardiac surgery: effects of avoiding cardiopulmonary bypass and the use of bilateral internal mammary arteries. Division of Nephrology. N Engl J Med 2000. Locker C. 347: 984–85. The effect of previous coronary-artery bypass surgery on the prognosis of patients with diabetes who have acute myocardial infarction. 2001 Copyright © 2001 All Rights Reserved 875 . Department of Surgery (M Winkler MD). She could not answer simple questions. University of Hannover School of Medicine. an abscess of the splenic parenchyma and fresh rupture of the capsule. Eventually. 18: 31–37. et al. et al. She had a long history of schizophrenia and lived in a psychiatric hospital. We did a laparotomy and a splenectomy. Kramer A. 69: 841–46. Lancet 1996.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue listening from where you left off, or restart the preview.