Sex dilferences in early and long-term results of

Transcription

1 THERPY ND PREVENTION PTC Sex dilferences in early and long-term results of coronary angioplasty in the NHLI PTC Registry MICHEL J. COWLEY, M.D., SUZNNE M. MULLIN, RN., M.P.H., SHERYL F. KELSEY, PH.D., KENNETH M. KENT, M.D., NDRES R. GRUENTZIG. M.D., KTHERINE M. DETRE, M.D., DR. P.H., ND EUGENE R. PSSMNI, M.D. STRCT To assess whether gender influenced the outcome of percutaneous transluminal coronary angioplasty (PTC), we analyzed data from the NHLI PTC Registry. Early results were compared in 705 women and 2374 men. Women were older (p <.01) and had more unstable angina (p <.01), and class 3 or 4 angina (p <.01). Men had more multivessel disease (p <.01), prior bypass surgery (p <.01), and abnormal left ventricular function (p <.05). Women had a lower angiographic success rate (60.3 vs 66.2%; p <.01) and had a lower clinical success rate (56.6% vs 62.2%; p <.01). More women had complications (27.2% vs 19.4%; p <.01), but overall frequency of major complications (death, myocardial infarction, emergency surgery) was not different (9.8% vs 9.3%). Women had a higher incidence of coronary dissection (p <.05) and higher in-hospital mortality (1.8% vs 0.7%; p <.01). PTC-related mortality was nearly six times higher in women (1.7% vs 0.3%; p <.001) and mortality with emergency surgery was more than five time higher (17.4% vs 3.2%; p <.001). Multivariate analysis indicated that female gender was an independent predictor for lower success (p <.05) and early mortality (p <.05) and was the only baseline predictor for PTC-related mortality. Late results in 2272 patients from centers with virtually complete follow-up of 1 year or longer (mean 18 months) showed comparable or better results in women than men. Men had higher rates of angiographic restenosis (36% vs 22%; p <.01), repeat PTC (18% vs 10%; p <.01), additional revascularization (27% vs 18%; p <.01), and cumulative mortality (2.2% vs 0.3%; p <.05), and frequency of symptomatic improvement similar to that in women. These NHLI Registry data indicate that PTC in women was associated with less favorable short-term outcome, lower initial success rate,and higher mortality rate than in men. However, longer-term results after PTC were comparable or better in women, with similar symptomatic improvement. lower rates of restenosis, and improved survival compared with men. Circulation 71, No. 1, 90-97, CORONRY artery bypass surgery is an effective means of myocardial revascularization that provides symptomatic improvement in most patients. Factors influencing outcome with coronary bypass surgery have been identified, and a number of reports indicate that less favorable results are achieved in women than in men.` These differences in results include higher operative mortality, lower graft patency rates, and less symptomatic improvement in women. Percutaneous transluminal coronary angioplasty (PTC), a newer, nonsurgical technique, is also a safe and effective From the Medical College of Virginia, Richmond; University of Pittsburgh Graduate School of Public Health, Pittsburgh; Georgetown University Medical Center, Washington, D.C.; Emory University School of Medicine, tlanta; and the Cardiac Diseases ranch, National Heart, Lung, and lood Institute, ethesda. ddress for correspondence: Michael J. Cowley, M.D., Medical College of Virginia, ox 59, MCV Station, Richmond, V Received June 1, 1984; revision accepted Oct. 11, Presented in part at the 56th Scientific Sessions of the merican Heart ssociation, naheim, method of myocardial revascularization in selected patients.9 12 However, the effects of gender on outcome with PTC have not been characterized. The NHLI PTC Registry has collected data on more than 3000 patients who have undergone PTC, and these data were analyzed to assess whether gender also influences results with PTC. Methods Data collected on 3079 patients enrolled in the NHLI PTC Registry through September 1982 were analyzed. The results include only first PTC procedures in these patients. dditional patients enrolled under special protocols (PTC in conjunction with thrombolysis in acute myocardial infarction, patients with cardiogenic shock or with recent myocardial infarction) were not included in this analysis. Extent of coronary artery disease was determined with 50% or greater diameter narrowing considered significant. ngina was graded according to the classification system of the Canadian Heart ssociation. 13 Unstable angina was defined as a change in the pattern of chest pain within 2 months of enrollment. ngiographic success of angioplasty was defined as 20% or greater reduction in luminal diameter narrow- CIRCULTION

2 ing. Clinical (short-term outcome) success was defined as angiographic success of all lesions treated without myocardial infarction, death, or bypass surgery during the subsequent hospital course. Definitions of complications from the PTC manual of operations were used and have been reported previously. ' 14 ll in-hospital deaths were reviewed by an independent external review committee to evaluate relation to the angioplasty procedure and were classified as related or probably not related to the procedure. Virtually complete follow-up information was obtained in a subset of 2272 patients from 65 clinical sites; each of these centers achieved 90% or greater follow-up (overall 98.5%) of at least 1 year (mean 18 months). Follow-up information was collected on the anniversary date of PTC and included vital status, presence and frequency of angina, hospitalizations for chest pain or myocardial infarction, and need for medication, repeat PTC, or coronary bypass surgery. Restenosis was defined as 30% or greater increase in stenosis from the initial post-ptc angiogram or loss of 50% or more of the initial gain in improvement of luminal diameter. Restenosis was assessed in a subset of 665 patients from 27 sites that had 75% or greater angiographic follow-up after successful PTC; angiographic follow-up was available in 557 of these 665 patients (84%). 15 Lesion location was defined as proximal or nonproximal according to the coronary anatomy nomenclature of the Coronary rtery Surgery Study.'16 The techniques of PTC have been reported previously. 0 standard dilatation catheter (3 mm balloon diameter) was the smallest size available during most of the registry experience and was used with similar frequency (>90%) in men and women. To evaluate the relationship of gender to outcome with PTC, baseline clinical and angiographic factors were examined as potential risk factors for success, complications, and mortality for each sex separately. Statistical analysis was performed with a chi square test to assess differences in categorical variables. Factors identified as significant for men or women were included along with sex as explanatory variables in a stepwise multivariate logistic regression model to predict outcome.1'7 Results Clinical characteristics. The clinical and angiographic characteristics of men and women enrolled in the NHLI Registry are listed in table 1. There were 705 women and 2374 men. The mean age of women was higher than that of men, and women had a higher frequency of severe angina (functional class 3 or 4) and a higher incidence of unstable angina. Women also had a longer median duration of angina, with a higher proportion having angina of more than 12 months duration; and more women had a history of hypertension than men. Men had a higher incidence of multivessel coronary disease than women, a higher frequency of prior bypass surgery, and a higher rate of left ventricular ejection fraction below 50%. The prevalence of prior myocardial infarction, significant left main disease, distribution of different PTC vessels attempted, and the frequency of multiple dilatation attempts were not different between men and women. Vol. 71, No. 1, January 1985 THERPY ND PREVENTION-PTc TLE 1 Clinical profiles of men and women in the NHLI PTC Registry No. of patients ge (mean) ngina ngina class (CHC) Unstable angina ngina duration Median <6 mo 6-12 mo >12 mo Hypertension Prior MI Prior bypass LVEF <50% Extent CD Single vessel Multivessel Left main PTC vessel Women yr mo C Men yr mo LD RC LCx Graft Left main Multiple attempts Mean % stenosis CD = coronary artery disease; CHC = Canadian Heart Classification; LVEF = left ventricular ejection fraction; LD = left anterior descending artery; LCx = circumflex artery; RC = right coronary artery; MI = myocardial infarction. Values are percentages except as indicated. Statistical comparisons: p <.05; p <.01; Cp <.001. ngioplasty results. The clinical results of angioplasty are shown in table 2. The success rate with coronary angioplasty was significantly different between men and women. Women had a lower angiographic success rate and a lower clinical (short-term outcome) success rate than men. In patients with single-vessel disease, men had a higher success rate than women; in the smaller number of patients with multivessel disease, success rate in men was also higher but the difference was not significant. Women had a lower rate of successful dilatation of the left anterior descending artery than men; the success rates of PTC of other vessels were similar in men and women. The most frequent reason for unsuccessful PTC was inability to pass the lesion. Women had a higher incidence of inability to pass the lesion, caused primar- - 91

3 COWLEY et al. TLE 2 Clinical results of PTC Women (%) Men (%) ngiographic success Clinical success Mean % stenosis change Unsuccessful result Inability to pass (failure to cannulate) (5.0) (2.0) Inability to dilate brupt reclosure Other failure Intimal tear ypass surgery Elective Emergency Statistical comparisons: p <.05; p <.01. ily by a higher incidence of inability to cannulate the PTC vessel. The incidence of inability to dilate the lesion once traversed, of abrupt reclosure of the vessel, and of other causes of failure were similar. Intimal tear associated with angioplasty was also more frequent in women than men. More women underwent bypass surgery after PTC; the incidence of elective surgery was significantly higher in women, whereas the incidence of emergency bypass surgery was similar in men and women. Similar proportions of men and women (86.4% and 86.7%) had either successful angioplasty or underwent subsequent bypass surgery. Complications. The distribution of complications reported in the NHLI PTC Registry is presented in table 3. Women had a higher incidence of any complication (which includes minor complications). Coronary dissection was more frequent in women. However, the overall frequency of a major complication (death, nonfatal myocardial infarction, or the need for emergency bypass surgery) was similar in men and women, and the incidence of nonfatal myocardial infarction and emergency bypass surgery was not different in men and women. In-hospital death occurred in 29 patients (0.9%). Hospital mortality rate was significantly higher in women (13/705, 1.8%) than in men (16/2394, 0.7%; p <.01). Ten of the 29 deaths were classified by the external review committee as probably not related to PTC, and nine of these occurred in men. Thus, when only PTC-related mortality is considered, the gender difference is more pronounced with a PTC-related mortality of 1.7% (12/705) in women and 0.3% (7/ 2374) in men (p <.001). The mortality rate with emergency surgery was also significantly higher in women (17.4%) than in men (3.2%; p <.001). Mortality rate with elective surgery after unsuccessful PTC was similar in men and women (1.7% vs 1.8%). Factors influencing success rate. Factors influencing clinical success rates in men and women are shown in table 4. Univariate factors associated with significantly lower success rates in men or women were multivessel disease (in men), class 3 or 4 angina (men), unstable angina (women), attempt in a vessel other than the left anterior descending artery (in men), lesion severity of 90% or greater narrowing (men and women), age 60 years or older (women), and angina duration 6 months or longer (men). Multivariate analysis of 2212 patients with complete data for these univariate factors identified six variables that were independently associated with significant reductions of success rate. Ranked in order of significance, these were lesion severity of 90% or greater narrowing, lesion calcification, nonproximal lesion, multivessel disease, long duration of angina, and female sex. Thus female sex remained a significant independent predictor of reduced success when adjusted for other baseline variables. Factors influencing mortality. aseline clinical characteristics identified as risk factors for hospital mortality in men and women are listed in table 5. Prior coronary bypass surgery was associated with a significantly higher mortality in men (2.5% vs 0.5%; p <.01); mortality in women with prior bypass surgery was also higher (5. 1% vs 1.7%) but the difference was not significant, probably because of the small number of women with prior bypass surgery. Mortality in patients TLE 3 Comparison of complications with PTC in men and women. Women (%) Men (%) Patients Prolonged angina MI Occlusion Dissection Spasm radycardia Hypotension Vent. fibrillation Peripheral arterial Major complications Death Nonfatal MI Emergency surgery PTC-related mortality 1.7 c 0.3 Emergency CG mortality 17.4 c 3.2 MI = myocardial infarction; CG = coronary artery bypass graft surgery. Statistical comparisons: p <.05; p <.01; Cp < CIRCULTION

4 THERPY ND PREVENTION-PTC TLE 4 aseline factors associated with reduced success rate Success rate Success rate Univariate factors men (%) women (%) Multivariate factors Lesion o90%c Lesion -90% vs <90% 55 vs 72c 48 vs 67c Lesion calcificationc MVD vs SVD 56 vs 65C 50 vs 58 Nonproximal lesion FC 3 or 4 vs FC 1 or 2 60 vs 68C 56 vs 60 MVD Non-LD vs LD attempt 56 vs 66c 54 vs 58 ngina >6 mo Distal vs proximal lesion 56 vs 66c 53 vs 59 Female gender Lesion calcium vs none 49 vs vs 65 Unstable vs stable angina 62 vs vs 62 ge >60 yr vs <60 yr 59 vs vs 61 ngina >6 mo vs <6 mo 59 vs vs 61 FC = functional class of angina; MVD multivessel disease; SVD single-vessel disease; LD = left anterior descending artery. Statistical comparisons: p <.05; p <.01; Cp <.001. with prior bypass surgery was not different between women and men. Women with abnormal ejection fraction (<50%) had a higher mortality than those with normal ejection fraction (5.9% vs 1.4%); ejection fraction did not influence mortality in men. In women mortality was higher with PTC attempts in the circumflex coronary artery than with attempts in other vessels (9.4% vs 1.5%; p <.05); in men mortality was higher with PTC attempts in the left main artery (3. 1 %) and with bypass graft attempt (3.3%) than with PTC in other vessels (0.5%). ge over 60 years was a significant risk factor for mortality in men (1.7% vs 0.4% in men <60 years of age); mortality was the same (1.8%) in women older than 60 years and in younger women. Long duration of angina (>12 months) was associated with a higher mortality in men (1.5% vs 0.4%) but not in women (2.3% vs 1.7%). Multivariate analysis performed in 2798 patients with complete data on these factors identified three independent predictors for in-hospital mortality. These TLE 5 Factors associated with early mortality in the NHLI Registry Univariate Multivariate Factor p value Sex Factor p value Gender <.01 F Prior CG <.01 Prior CG <.01 M Female gender <.01 ge >60 <.01 M ge >60 <.05 ngina >12 mo <.05 M LCx attempt <.05 F Left main attempt <.05 M Graft attempt <.05 M LVEF <50% <.05 F CG = coronary artery bypass graft surgery; LCx = left circumflex artery; LVEF = left ventricular ejection fraction. Vol. 71, No. 1, January 1985 were prior bypass surgery, female sex, and age over 60 years. Thus, when adjusted for other baseline variables, female gender remained a highly significant risk factor for increased mortalit) with PTC. Female gender was the only predictor for PTC-related mortality. Influence of body size on outcome. nalysis of outcome as a function of body size was also performed with the parameters of height, weight, and body surface area. No relationship between body size and success or complications was identified. significant inverse linear relation was identified between height and mortality rate in men, with higher mortality in shorter men (p <.05 for chi square test for linearity). The mortality rate in the group of men under 165 cm (1.7%) was comparable to overall mortality (1.8%) in women. No relationship between body size and mortality was identified in women. Long-term follow-up. Follow-up data were available in 2272 patients: 1418 at 1 year, 606 at 2 years, 191 at 3 years, and 57 at 4 years after PTC. Mean follow-up for the group was 18 months. PTC was successful in 1397 patients in the follow-up cohort (61.5%); the follow-up group with successful PTC included 305 women and 1092 men. The duration of follow-up in the successful group was 1 year in 906, 2 years in 349, 3 years in 108, and 4 years in 34 patients. Follow-up information was available in 1382 (98.9%) of these patients. The clinical results during follow-up in men and women with initially successful PTC are listed in tables 6 and 7. During the follow-up period, 75.3% of patients did not require additional revascularization. The incidence of additional revascularization was significantly higher in men than in women (27.0% vs 17.9%), and repeat PTC was performed more often in men than in women (17.5% vs 9.9%). The frequen- 93

5 COWLEY et al. TLE 6 Follow-up status after successful PTC in the NHLI Registry Men Women p value No. of patients Symptomatic improvement NS Event-free survival <.01 Event-free, no symptoms NS Repeat PTC <.05 Repeat PTC or CG <.01 CG NS MI NS Death <.05 CG = cornary artery bypass graft surgery; MI myocardial infarction. Values are percentages except No. of patients. Includes additional revascularization with CG or repeat PTC. cy of coronary artery bypass surgery during follow-up was not significantly different in men and women. ngiographic restenosis, determined from the subset of centers with high angiographic follow-up rates, was significantly more common in men (163/450, 36.2%) than in women (24/107, 22.4%; p <.01), as previously reported.'5 etween the time of PTC and final follow-up contact, chest pain was reported as improved or eliminated in 91.2% of patients (including additional revascularization), and the proportion with improvement was similar in men (91.7%) and women (89.8%). Event-free survival (no additional revascularization, myocardial infarction, or death) was higher in women (79.7%) than in men (69.0%). Similar pro- 94 TLE 7 Cumulative event rates after successful PTC Death portions of men and women were event-free and asymptomatic (51.2% vs 52.7%). The frequency of myocardial infarction during follow-up was 4.0% in both men and women. Death occurred in 25 patients (1.8%) during follow-up; late mortality was significantly higher in men (2.2%) than in women (0.3%; p <.05). The cumulative event rates (death; myocardial infarction; death or myocardial infarction; repeat PTC, bypass surgery, death, or myocardial infarction) in the follow-up cohort of men and women with successful PTC are listed in table 7. The cumulative mortality was significantly higher in men at 1 year and at 2 years. The cumulative rates of myocardial infarction were not significantly different. The event rate for death or nonfatal myocardial infarction was higher in men at 6 months but not at 1 and 2 years. The cumulative combined event rate (repeat PTC, bypass surgery, death, or myocardial infarction) was significantly higher in men throughout the follow-up period. Factors influencing late survival after PTC were similar for the total follow-up cohort and for the group with initially successful PTC (table 8). Factors associated with increased late mortality by univariate analysis were male gender (2.2% vs 0.3%), history of myocardial infarction before PTC (3.6% vs 1.3%), presence of left main disease (12. 1 %, multivessel 4.0%, single vessel 1.0%), left main attempt (17.9% vs 1.5%), initial functional class 3 or 4 angina (2. 1% vs 0. 8%), and history of hypertension (3. 1% vs 1. 1%). y multivariate analysis, left main disease, male gen- Male Female Male Female n % SE n % SE n % SE n % SE 30 days /2 yr yr /2 yr yr Death/MI MI Repeat PTC/CG/death/MI Male Female Male Female n % SE n % SE n % SE n % SE 30 days /2 yr gc 1 yr c 11/2 yr yr MI = myocardial infarction: CG coronary artery bypass graft surgery. < 05; p <.01: Cp <.001. CIRCULTION

6 TLE 8 Factors associated with late mortality after PTC in the NHLI Registry p p Univariate value Multivariate value Left main disease <.01 Left main disease <.01 Multivessel disease <.01 Male gender <.05 Male gender <.05 Class 3 or 4 angina <.05 Prior MI <.01 Hypertension <.05 Left main attempt <.01 Multivessel disease.10 Class 3 or 4 angina <.05 Smoking history.10 Hypertension <.05 LVEF <50% <.05 Smoking history <.05 MI = myocardial infarction; LVEF = left ventricular ejection fraction. der, functional class 3 or 4 angina, and history of hypertension were identified as significant independent predictors of late mortality after PTC. Discussion These data from the NHLI PTC Registry demonstrate significant differences in outcome between men and women undergoing PTC and identify factors influencing outcome by sex and for the total population. Women had a lower success rate and higher complication and mortality rates than men. Gender differences in outcome have also been identified in numerous reports of myocardial revascularization with coronary bypass surgery. 1-' The NHLI Registry data also show significant differences in baseline clinical and angiographic characteristics between men and women, which may contribute to these differences in results. Women were older as a group and had more unstable angina and more severe angina (class 3 or 4). However, women had less extensive coronary artery disease, fewer patients with abnormal left ventricular function, and fewer with prior coronary bypass surgery. Similar differences in baseline characteristics between men and women have been reported in patient populations undergoing coronary bypass surgery.5 6. Multivariate analysis of factors influencing outcome allows adjustment for differences in baseline variables. The lower success rate with PTC in women was highly significant by univariate analysis, and female gender remained a significant independent predictor of reduced success when adjusted for other baseline variables. The most important predictors of lower success rate were coronary anatomic factors (lesion severity >90%, lesion calcification, and nonproximal lesions), which were each highly significant. The less favorable results associated with these angiographic factors were antici- Vol. 71. No. 1, January 1985 THERPY ND PREVENTION-PTc pated in the original guidelines for patient selection for PTC'0, 1i and were most likely caused by technical limitations of the available angioplasty catheter systems in reaching, traversing, and dilating these lesions. Other significant factors associated with lower success rates were multivessel disease and long duration of angina. These data represent experience with PTC during the initial 4 years of its clinical application. Since the close of the NHLI Registry, important technical improvements in catheter systems (smaller profile catheters, high-pressure balloons, steerable dilatation systems) have been introduced and success rates in excess of 80% to 90% are currently achieved at many centers. 19 lthough female gender and these other predictors of reduced success identified in the registry may have less influence on current results with better technology and higher initial success rates, they were significant factors in the registry experience and may remain important determinants of a successful outcome with PTC. The higher incidence of bypass surgery in women reflects the lower primary success rate with PTC. The higher frequency of overall complications in women was primarily caused by a higher incidence of certain minor complications, particularly bradycardia and hypotension. The higher incidence of coronary dissection (as well as the higher frequency of intimal tear) with angioplasty in women may have been caused by a number of factors, including differences in plaque composition, structural characteristics of the vessel wall, or more likely to differences in vessel size. lthough data on vessel size are not available from the registry, body size has been shown to correlate with coronary artery size.20 lthough women are smaller than men, the same balloon catheter size (3 mm diameter) was used in most men and women. The prevalence of other important nonfatal complications (myocardial infarction, coronary occlusion, coronary spasm, need for emergency bypass surgery) were not influenced by gender. In addition, the overall frequency of major complications (death, nonfatal myocardial infarction, need for emergency bypass surgery) was not different in men and women. However, women undergoing PTC had a significantly higher in-hospital mortality than men. similar increase in mortality with myocardial revascularization in women has been reported with coronary bypass surgery.'-',21 Univariate factors in women that were associated with a significant risk of mortality with PTC were left ventricular ejection fraction under 50% and PTC attempt in the left circumflex artery. 95

7 COWLEY et al. These factors were not significant in men. Women had a significantly higher mortality than men in the age group under 60 years. Mortality in men increased significantly beyond 60 years and was not different than the mortality in women over 60 years or the overall mortality in women. similar relationship among age, gender, and mortality has been reported with bypass surgery. 6 8,21 PTC-related mortality showed even more striking differences by gender, with an almost sixfold higher mortality in women. Emergency surgery was performed for complications of PTC in most of these patients who died, and the mortality with emergency surgery was more than five time higher in women than in men. The major difference in mortality by gender therefore occurred predominantly in the group who underwent emergency surgery. Thus the known increase in operative risk with bypass surgery in women may be an important contributing factor in the higher hospital mortality encountered with PTC in women and may be amplified when emergency operation is necessary. However, the operative risk with elective surgery after unsuccessful PTC was similar in men and women. Multivariate predictors for increased mortality in the registry, ranked in order of importance, were prior bypass surgery, female gender, and age over 60 years. lthough women were older. age over 60 years was a significant risk factor only in men, as was prior bypass surgery. Similarly. a relationship between body size and mortality was identified only in men. Several studies of coronary bypass surgery suggest that differences in body size are a major factor in the higher operative mortality encountered in women.8 21 These studies have also reported similar mortality in women and in smaller men. The higher mortality in the registry in smaller men. which was similar to the overall mortality in women. is consistent with this observation. However, body size was not found to be an independent predictor of mortality with PTC in the registry. In contrast to the less favorable short-term outcome with PTC in women, the longer-term results with PTC were comparable or better in women than in men. During the mean follow-up period of 18 months after successful angioplasty, men had a significantly higher incidence of restenosis'l and a higher frequency of repeat PTC and additional revascularization (repeat PTC or bypass surgery). Similar proportions of men and women were event-free and asymptomatic. Including additional revascularization, similar proportions reported symptomatic improvement at last follow-up and the incidence of myocardial infarction was not different during follow-up. These results differ with several reports of long-term results after bypass surgery suggesting that women have less symptomatic improvement and lower patency rates than men.'.' Men also had a lower event-free survival and a higher cumulative mortality at 1 and 2 years after PTC than women. In contrast, follow-up studies after bypass surgery suggest that long-term survival is comparable in men and women. 4 7 X.8-2 It is well known that men have a higher mortality than women in the general population. y applying the 1979 U.S. sex- and 5 year age-specific mortality rates for ischemic heart disease to this PTC population, one female and five male deaths would be expected in 1 year 3; one female death and 24 male deaths occurred. Thus male mortality exceeded the general population rate by four times, and mortality in women was the same as in the general population. Since the PTC population includes only men and women with coronary heart disease, which is known to be less prevalent in women, the low number of female deaths appears even more striking and suggests that the differences in survival after PTC are not attributable to intrinsic differences in prognosis and survival by gender. Multivariate predictors of late mortality after PTC were left main disease, male gender, class 3 or 4 angina, and history of hypertension. Of these factors, fewer men had class 3 or 4 angina and hypertension and a similar proportion of men and women had left main disease. The higher late mortality in men is therefore not clearly explained by differences in baseline varibles but may be partly related to their higher prevalence of multivessel disease, which approached but did not reach significance as an independent predictor of late mortality. Whether this survival advantage in women after PTC will persist will require longerterm follow-up. In summary, these data from the NHLI PTC Registry demonstrate that PTC in women is associated with a significantly lower success rate and a higher in-hospital mortality rate than in men. However, during follow-up after successful PTC, women have comparable symptomatic improvement, a lower incidence of additional revascularization and restenosis, and improved survival compared with men. References 1. olooki H, Vargas J, Green R. Kaiser G. Ghahramani : Results of direct coronary artery surgery in women. J Thorac Cardiovasc Surg 69: Sheldon WC, Rincon G, Pichard D. Razavi M. Cheanvechai C, Loop FC: Surgical treatment of coronary artery disease: pure graft operations with study of 751 patients followed 3-7 years. Prog Cardiovasc Dis 18: 237, Golding LR. Groves LK: Results of coronarv artery surgery in women. Cleve Clin Q 43: CIRCULTION

8 THERPY ND PREVENTION-PTC 4. Tyras DH, arner H, Kaiser GC, Codd JE, Laks H, Willman VL: Myocardial revascularization in women. nn Thorac Surg 25: 449, Kennedy JW, Kaiser GC, Fisher LD, Fritz JK, Myers JW, Mudd JG, Ryan TJ: Clinical and angiographic predictors of operative mortality from the Collaborative Study in Coronary rtery Surgery (CSS). Circulation 63: 793, Kennedy JW, Kaiser GC, Fisher LD, Maynard C, Fritz JK, Myers W, Mudd JG, Ryan TJ, Coggin J: Multivariate discriminant analysis of the clinical and angiographic predictor of operative mortality from the Collaborative Study in Coronary rtery Surgery (CSS). J Thorac Cardiovasc Surg 80: 876, Douglas JS Jr, King S, Jones EL, Craver JM, radford JM, Hatcher CR Jr: Reduced efficacy of coronary bypass surgery in women. Circulation 64 (suppl II): , Loop FD, Golding LR, MacMillan JP, Cosgrove DM, Lytle W, Sheldon WC: Coronary artery surgery in women compared with men: analyses of risks and long-term results. J m Coll Cardiol 1: 383, Gruentzig R, Senning, Siegenthaler WE: Nonoperative dilatation of coronary artery stenosis: percutaneous transluminal coronary angioplasty. N Engl J Med 301: 61, Kent KM, entivoglio LG, lock PC, Cowley MJ, Dorros G, Gosselin J, Gruentzig, Myler RK, Simpson J, Stertzer SH, Williams DO, Fisher L, Gillespie MJ, Mullin SM, Mock M: Percutaneous transluminal coronary angioplasty: report from the NHLI Registry. m J Cardiol 49: 2011, Dorros G, Cowley MJ, Simpson J, entivoglio LG, lock PC, ourassa M, Detre K, Gosselin J, Gruentzig, Kelsey SF, Kent KM, Mock M, Mullin SM, Myler RK, Passamani ER, Stertzer SH, Williams DO: Percutaneous transluminal coronary angioplasty: report of complications from the NHLI PTC Registry. Circulation 67: 723, Kent KM, entivoglio LG, lock PC, ourassa MG, Cowley MJ, Dorros G, Detre K, Gosselin J, Gruentezig, Kelsey SF, Mock M, Mullin SM, Passamani E, Myler RK, Simpson J, Stertzer SH, VanRaden M, Williams DO: Long-term efficacy of percutaneous transluminal coronary angioplasty (PTC): report from NHLI- PTC Registry. m J Cardiol 53: 27C, Campeau: Grading of angina pectoris. Circulation 54: 522, Cowley MJ, Dorros G, Kelsey SK, VanRaden MJ, Detre KM: cute coronary events associated with percutaneous transluminal coronary angioplasty: NHLI PTC Registry experience. m J Cardiol 53: 12C, Holmes DR Jr, Vlietstra RE, Smith HC, Vetrovec GW, Kent KM, Cowley MJ, Faxon DP, Gruentzig R, Kelsey SF, Detre KM, VanRaden MJ, Mock M: Restenosis after percutaneous transluminal coronary angioplasty (PTC): a report from the PTC Registry of the National Heart, Lung, and lood Institute. m J Cardiol 53: 77C, Principal Investigators and their ssociates: National Heart, Lung and lood Institute Coronary rtery Surgery Study. Circulation 63 (suppl 1): I-15, Schlesselman JJ: Multivariate analysis. In Schlesselman JJ: Casecontrol studies: design, conduct, analysis. Oxford, 1982, Oxford University Press, p Levy RI, Mock M, Willman VL, Frommer PL: Percutaneous transluminal coronary angioplasty. N Engl J Med 301: 101, Meier, Gruentzig R: Learning curve for percutaneous transluminal coronary angioplasty: skill, technology or patient selection. m J Cardiol 53: 65C, Roberts CS, Roberts WC: Cross-sectional area of the proximal portions of the three major epicardial coronary arteries in 98 necropsy patients with different coronary events: relationship to heart weight, age and sex. Circulation 62: 953, Fisher LD, Kennedy JW, Davis K, Maynard C, Fritz JK, Kaiser G, Myers WO: ssociation of sex, physical size, and operative mortality after coronary artery bypass in the Coronary rtery Surgery Study (CSS). J Thorac Cardiovasc Surg 84: 334, Killen D, Reed W, rnold M, McCallister D, ell HH: Coronary artery bypass in women: long-term survival. nn Thorac Surg 34: 559, Vital Statistics of the U.S : General mortality. Vol II, section 1, part, US DHHS, Public Health Service, p 21 Vol. 71, No. 1, January