HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Johnson, R. G. Articles by Weintraub, R. M. Search for Related Content PubMed PubMed Citation Articles by Johnson, R. G. Articles by Weintraub, R. M.

Ann Thorac Surg 1995;59:1391-1396
© 1995 The Society of Thoracic Surgeons

CABG After Successful PTCA: A Case-Control Study

Division of Cardiothoracic Surgery, Department of Surgery, and Cardiovascular Division, Department of Medicine, Beth Israel Hospital and Harvard Medical School, Boston, Massachusetts

	Abstract

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
We sought characteristics predictive of the need for operative revascularization subsequent to a successful coronary angioplasty. Through June 1993, 128 patients who had successful percutaneous transluminal coronary angioplasty (PTCA) between January 1982 and March 1989 required subsequent coronary artery bypass grafting (CABG) at our hospital. These cases were matched with 128 controls who had a successful PTCA but did not require subsequent CABG. Controls were matched to cases by the date of their initial PTCA. Before initial PTCA there were no differences between the cases and controls in terms of age, sex, prior myocardial infarction, ejection fraction, duration of anginal symptoms, hypertension, hyperlipidemia, family history, or obesity (all not significant). A greater number of cases had diabetes (35 versus 18; p = 0.009). Angiography before initial PTCA revealed that cases had a greater mean number of total lesions (4.1 versus 3.3; p = 0.002) and a higher incidence of left anterior descending and circumflex artery stenoses of 70% or greater (98 versus 75 and 57 versus 34, respectively; p = 0.006). The mean number of lesions successfully dilated was greater in cases (2.4 versus 1.7; p = 0.0001). Cases had CABG at a mean interval of 16.7 ± 23 months. There were 17 late deaths among cases and 9 among the controls at a mean of 38.6 ± 30 months. The survival probability at 5 years was 94.5% for controls and 87.9% for cases (p = 0.048). Initial revascularization by PTCA is followed by CABG at a brief interval in a subset of patients who have markers of more severe disease than do patients who do not require early CABG. Patients treated with PTCA and CABG have a poorer long-term survival. The method of initial revascularization should be considered carefully as markers of more severe disease may indicate primary CABG and avoidance of an initial PTCA.

	Introduction

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
See also page 1396.

Coronary revascularization procedures have increased in number steadily over the past decade as manifest by the aggregate number of percutaneous transluminal coronary angioplasties (PTCAs) and coronary artery bypass graft (CABG) procedures. The shifting proportions of these two procedures has been guided by a drive for less invasive treatment whenever possible, with operative revascularization increasingly being reserved for patients who are ``not candidates'' for further percutaneous revascularization. Only one randomized trial comparing percutaneous revascularization to medical therapy has been published [1], and few prospective trials have compared ``single-vessel'' or ``double-vessel'' disease treated by operation versus angioplasty [2]. Nevertheless, a number of trials comparing percutaneous to operative revascularization for multivessel disease have been reported [2–4], but these studies were not designed to provide insight into the decision process that selects PTCA as the initial method for revascularization.

Twenty-three to 41% of patients having successful angioplasty for multivessel disease in the randomized trials required subsequent angioplasties over intervals as short as 1 year. Eighteen to 20% of patients with multivessel disease treated with initial PTCA required an operation for revascularization (CABG) within 12 months after an initial, successful PTCA. It is not clear whether PTCA might have been obviated in those patients who required early operative revascularization after a successful PTCA.

In this study, we asked whether there were factors that identified patients who required early operation after an initial successful balloon angioplasty. We tested the null hypothesis that no pre-PTCA characteristics predict the need or timing for subsequent CABG. It was anticipated that, even if no predictive characteristics were identified, this study would provide data on the frequency and interval to CABG, the change in condition before CABG, and the overall survival after multiple procedures. We then hoped to infer whether the therapeutic decision process is appropriate and to suggest any recommendations for changing that selection process.

	Material and Methods

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
Population Studied
In review of operations performed at our hospital from January 1982 through June 1993, there were 128 consecutive patients who had successful PTCA between January 1982 and March 1989 but required a subsequent CABG. These cases were compared with controls consisting of patients who had PTCA but no subsequent CABG over the same time interval. During the period of operations reviewed (138 months) 2,916 CABG procedures were performed at our hospital. Over the period of PTCA procedures (87 months) 2,004 angioplasties were performed at our institution. A ``successful'' PTCA was defined as dilation of an intended coronary stenosis or stenoses such that the residual luminal narrowing was less than 40% and unassociated with a flow-limiting dissection or the need for immediate operative coronary revascularization.

Patients were matched solely by the date of angioplasty. Prospective controls who had emergency operations associated with their initial PTCA (35 patients), those who had ``unsuccessful'' angioplasty, and those who had prior CABG were excluded from the controls. If no control PTCAs were available on the date of a case angioplasty or if a potential control patient met exclusionary criteria, the next closest PTCA was matched. In the event of one or more control choices on a given case-date, the match was selected randomly.

Data Collection
Case and control charts were reviewed for the following characteristics before initial PTCA: age, sex, prior myocardial infarction, diabetes, family history of coronary disease, obesity, smoking, hyperlipidemia, and hypertension. Each catheterization report was reviewed for left ventricular ejection fraction, number and location of lesions, percent stenosis of each lesion, the number of lesions dilated, and PTCA complications. Any subsequent admission, myocardial infarction, or death was recorded for all patients. For cases the date of CABG, the number of vessels grafted, associated procedures, postoperative complications, and subsequent revascularizations also were recorded.

The risk of operative revascularization was estimated retrospectively for all patients using a risk assessment tool (Riskmaster; Health Data Research, Inc, Portland, OR) for cases and controls. This estimation was performed again for cases using data contemporary with their CABG. The data required for these determinations included age, sex, body surface area, obesity, diabetes, hypertension, chronic obstructive pulmonary disease, congestive heart failure (or ejection fraction), severe pulmonary vascular disease, prior cerebrovascular accident, prior myocardial infarction (with temporal association), hemodynamic status, prior operation, failed PTCA, left main stenosis (>50%), renal failure, and pulmonary artery pressure.

Surviving cases and controls were followed by telephone interviews. Survivors were asked about the decision for initial PTCA and current symptoms of angina, shortness of breath, exercise limitations, and the use of nitroglycerin. When sufficient data could be ascertained the New York Heart Association class was determined (12.4% of cases and 10.4% of controls could not have their New York Heart Association class definitively determined).

Data for hospital charges were available and tabulated for cases and controls hospitalized after 1983. One hundred twenty in each group had such data available. Evaluation of our institution's Medicare audited data from fiscal years 1987, 1988, and 1989 reveals a consistent cost/charge relationship for PTCA, CABG, and CABG with a diagnostic catheterization. The approximate cost/charge ratio over this period for these procedures was 0.66.

Statistical Methods
Count data were compared by ² analysis. Means were compared by two-tailed Student's t tests and are expressed as ± one standard deviation. Survival curves of equal-sized groups were constructed and analyzed by Mantel-Cox comparison, and for unequal samples by generalized Wilcoxon comparison. We modeled the occurrence of CABG (cases) as a function of baseline data (before initial PTCA) found to be significantly different from controls by univariate analysis (p < 0.01). These data were analyzed by stepwise multivariate logistic regression analysis (True Epistat; Epistat Services, Richardson, TX).

	Results

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
Before Initial PTCA
There were no differences in the mean age, the sex split, prior myocardial infarction history, ejection fraction, duration of anginal symptoms, hypertension, hyperlipidemia, obesity, or family history of coronary disease between the cases and controls. A significantly greater number of cases had diabetes (35 versus 18; p = 0.009). A comparison of characteristics before PTCA is shown in Table 1.

View larger version (44K): [in this window] [in a new window]   Fig 1. . Percentage of patients with one-, two-, and three-vessel disease (VD) in each group is contrasted with the percentage of patients who had various numbers of coronary lesions 70% described.

The baseline operative risk assessment of cases showed a 2.8% mean predicted operative risk compared with the controls' mean of 2.1% (p = not significant).

After Initial PTCA
CASES.
The 128 cases had their CABG performed at a mean of 16.7 months (range, 0 to 116 months) after a mean of 1.7 angioplasties. Sixty-three of the 128 patients had only the initial, single PTCA before CABG at a mean interval of 13.7 months. The other 65 patients had 92 additional PTCAs before operation. Twenty-two patients (17.2%) had CABG performed within 30 days of initial PTCA. Figure 2 shows that more than 80% of patients with four or more coronary stenoses of 70% or greater required operation within 1 year.

View larger version (19K): [in this window] [in a new window]   Fig 2. . Survival free of coronary artery bypass grafting (CABG) for cases according to the number of coronary lesions with a greater than 70% stenosis. Nearly all of the cases with 4 or more coronary stenoses 70% required their operation within 1 year. (PTCA = percutaneous transluminal coronary angioplasty.)

At the time of CABG the estimated operative risk assessment predicted a mean mortality of 4.0%, which was significantly higher than the 2.8% before initial PTCA (p = 0.04). There were three deaths (2.3%) within 30 days or in-hospital after CABG. Six patients (4.7%) had perioperative myocardial infarctions and there were no cerebrovascular accidents.

Revascularization procedures after CABG were required in 12 cases (9.4%). Eight patients had 10 PTCAs, and 4 had a second CABG at a mean interval of 69.7 ± 44 months. One additional patient had a third CABG 5 months after the second one. No further revascularization procedures were performed after CABG in 113 (88.3%).

CONTROLS.
Twenty-four controls (18.8%) had a second PTCA at a mean interval of 11 months. Over the entire period of follow-up the mean number of angioplasties performed in these patients (after their initial one) was 0.25/patient. Over the period of follow-up, severe morbid events occurred in 56 (43.7%) of the controls, including eight myocardial infarctions and nine deaths (one within 30 days of PTCA).

Long-Term Follow-up
The mean survival from initial PTCA was 71.1 ± 27 months for cases and 69.7 ± 28 months for controls. There were 17 late case deaths and 9 control deaths at a mean of 38.6 ± 30 months after initial PTCA. Survival data are shown in Figure 3. The probability of 5-year survival was greater for controls (94.5%) than for cases (87.9%; p = 0.048). At the time of initial PTCA the cases who experienced late death after CABG were older (63.9 versus 59.3; p = 0.087) and had lower mean LVEF (0.52 versus 0.63; p = 0.017). There was also a statistically significant difference in the estimated operative mortality that was calculated before initial PTCA in those who experienced late death (8.1% versus 1.99%; p = 0.005).

View larger version (17K): [in this window] [in a new window]   Fig 3. . Survival from initial percutaneous transluminal coronary angioplasty (PTCA) for the patients in both groups (cases had subsequent coronary artery bypass grafting and controls did not).

Multiple Logistic Regression Analysis of the Occurrence of Subsequent CABG
Among the baseline variables entered (diabetes, number of lesions, number of lesions with a greater than 70% stenosis and location of lesions), only diabetes (p = 0.042) and extent of coronary disease (p = 0.007) were found to be predictive of the need for subsequent CABG. The total number of coronary lesions described, regardless of severity, and the total number of lesions with a greater than 70% narrowing were too highly colinear to allow their simultaneous evaluation in the final regression model. Each alone, with diabetes, was highly predictive of an increased risk of subsequent CABG, but the number of lesions with a greater than 70% narrowing was slightly more predictive (odds ratio for ``number of lesions 70%'' = 1.469 [95% confidence interval = 1.193–1.809] versus ``number of any lesions'' = 1.193 [95% confidence interval = 1.050–1.355]) (Table 3). The location of lesions was not selected as adding predictive value (p = not significant).

Hospital Charges
The mean hospital charges per patient for their initial PTCA were not significantly different between controls and cases. The mean charges were $13,085 ± 6,775 for cases and $11,914 ± 5,815 for controls (p = 0.175). The mean charges for a CABG admission were $30,247 ± 14,820 per case. For PTCA plus CABG during the same admission (n = 32), the mean charges were $44,384 ± 41,635 per case (median = $35,056). The total charges for pre-CABG PTCAs among cases were $2.8 million or a mean of $22,095/patient.

	Comment

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
When diagnostic coronary angiography is performed for clinical ischemic cardiac disease the therapeutic choices include two methods of coronary revascularization, either operative (CABG) or catheter-based (PTCA). The best means of making this therapeutic selection has not been defined. A number of retrospective studies have looked at the results of that selection process. They have shown that operative revascularization was the initial choice in older patients with longer anginal histories, more associated medical illnesses, lower ejection fractions, more prior infarctions, and more multivessel disease [5–8]. Given the pre-revascularization selection bias for CABG in higher risk patients, the similarity of postprocedure morbidity and mortality in patients having either PTCA or CABG was remarkable [8, 9]. Indeed, Akins and associates' [6] comparison showed no difference in short-term mortality. That group further demonstrated that although advanced age and decreased ejection fraction were associated with a survival disadvantage (also characteristics of the CABG patients), the 5-year survival in patients treated initially with operative revascularization was not strikingly different from that of patients treated by PTCA (92% versus 96%, respectively) [6]. Although these reports did not provide insight as to the decision process they certainly demonstrated that the revascularization technique chosen, in practice, for higher risk patients was CABG [10], whereas the lower risk patients had PTCA, with similar early and late results.

The question of how lower risk patients do with CABG compared with PTCA also has been examined retrospectively. A subset of Coronary Artery Surgery Study Registry patients were considered to have been ideal PTCA candidates (2.8% of the 9,369 patients) based primarily on anatomic considerations. The operative mortality in these patients was 0.4% (1/261), with a perioperative infarction incidence of 2.3% [11]. Both figures compared favorably with published results for PTCA. Hochberg and colleagues [12] retrospectively matched patients who had operation for disease that was anatomically suitable for PTCA (single-or double-vessel disease) with patients who had PTCA. The hospital mortality for CABG was 1% (1/125) compared with 3% 30-day PTCA mortality and a 10% emergency CABG rate. Over 3 years of follow-up another 9% of the PTCA patients required a CABG [12]. These data demonstrated that patients who are selected for PTCA have a similar outcome to those who have CABG, if anatomically matched according to coronary disease. In making the clinical choice, however, the morbidity [3, 4] and expense [13] of CABG necessitate a better than equal outcome if it is to be selected over PTCA as the initial revascularization choice for any group of patients.

Published prospective trials comparing CABG and PTCA for initial revascularization [2–4] of multivessel disease did not show a significant difference in survival, in spite of the fact that the PTCA group required more subsequent revascularizations and the CABG group had more procedure-related infarcts. Nevertheless, if operative revascularization is required at some early interval after PTCA, and if such patients have a higher morbidity with subsequent CABG, then there are economic and medical arguments that urge triage of such patients to CABG for their initial revascularization.

In this retrospective series we examined only patients who had an initial, successful revascularization by PTCA. In accord with the findings of prior retrospective studies the patients in this study were relatively young and healthy. In spite of the many similarities between the groups (those who never had CABG and those who ultimately required CABG) there were demonstrable differences. The patients who had CABG had a significantly greater incidence of diabetes and significantly more coronary disease.

Assessment of anatomic disease by the traditional one-, two-, or three-vessel disease categories was not as discerning as the total number of lesions (regardless of severity) or the number of lesions more than 70% narrowed. These indices of diffuse or multicentric disease more accurately predicted a short interval to operative revascularization. Importantly, the operative mortality, as estimated by a multifactorial risk assessment program, increased significantly from what it was before the initial revascularization. Furthermore, the long-term survival for those patients who subsequently had CABG differed significantly from that of the PTCA patients who did not require CABG. It is likely that either the mortality of low-risk patients was increased by a delay in operation or the patients in the CABG group were simply at higher risk due to observable factors such as multivessel disease and diabetes. Either explanation suggests a need to identify patients before an initial PTCA who are going to require early operation. There are certainly no data implying that initial PTCA benefited patients who came to subsequent CABG, and indeed the reverse is suggested.

One of the major flaws of this study is the fact that the patients reviewed were treated by an initial PTCA from 1982 to 1989. The judgments made then might differ from contemporary ones, and a broader array of angioplasty techniques exist today. We bear this in mind and ask the readers to do the same. We did not analyze closely the reason for timing of nor decision for operative revascularization. No single-reviewer assessment of the patients' cineangiograms was performed. An objective, independent review of the cineangiograms might have been even more revealing in terms of anatomic stratification of these patients, but in this retrospective study we chose to use angiographic interpretations that were contemporary with the therapeutic decisions. Finally, we only sampled the PTCA group that did not require CABG by obtaining matched controls for the cases from a larger pool of patients who underwent angioplasty. Although we do not doubt the controls' representativeness, our methods precluded our determining the exact incidence of CABG subsequent to successful PTCA.

The factors that we identified can be used to select for initial revascularization by operation, avoiding the cost and morbidity of one or more angioplasties. Diabetes and the number of total lesions, or the number and location of lesions greater than 70% narrowed, should be considered in light of other factors (such as increased age and decreased ejection fraction) in choosing the method of initial revascularization. The traditional ``one-, two-, three-vessel'' disease classification is a poor discriminator in the selection of the appropriate initial revascularization technique. Stratification of patients by predictive criteria should be studied in a prospective manner to validate their ability to improve survival and reduce the total cost of coronary revascularization.

	Acknowledgments

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
Supported in part by a grant from the New England Foundation.

We gratefully acknowledge the contributions of Harold G. Roberts, MD, and Mallory Van Horn, BS, in the preparation of the manuscript and the follow-up of the patients in the study.

	Footnotes

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 
Presented at the Thirty-first Annual Meeting of The Society of Thoracic Surgeons, Palm Springs, CA, Jan 30-Feb 1, 1995.

Address reprint requests to Dr Johnson, Division of Cardiothoracic Surgery, Beth Israel Hospital, 330 Brookline Ave, Boston, MA 02215.

	References

Top Footnotes Abstract Introduction Material and Methods Results Comment Acknowledgments References

 

1. Parisi AF, Folland ED, Hartigan P. A comparison of angioplasty with medical therapy in the treatment of single-vessel coronary artery disease. Veterans Affairs ACME Investigators. N Engl J Med 1992;326:10–6.[Abstract]
2. Hampton JR, RITA Trial Participants. Coronary angioplasty versus coronary artery bypass surgery: the Randomised Intervention Treatment of Angina (RITA) trial. Lancet 1993;341:573–80.[Medline]
3. Hamm CW, Reimers J, Ischinger T, et al. A randomized study of coronary angioplasty compared with bypass surgery in patients with symptomatic multivessel coronary disease. N Engl J Med 1994;331:1037–43.[Abstract/Free Full Text]
4. King SB, Lembo NJ, Weintraub WS, et al. A randomized trial comparing coronary angioplasty with coronary bypass surgery. N Engl J Med 1994;331:1044–50.[Abstract/Free Full Text]
5. Jones EL, Murphy DA, Craver JM. Comparison of coronary artery bypass surgery and percutaneous transluminal coronary angioplasty including surgery for failed angioplasty. Am Heart J 1984;107:830–5.[Medline]
6. Akins CW, Block PC, Palacios IF, Gold HK, Carroll DL, Grunkemeier GL. Comparison of coronary artery bypass grafting and percutaneous transluminal coronary angioplasty as initial treatment strategies. Ann Thorac Surg 1989;47:507–15.[Abstract]
7. Davis PK, Parasandola SA, Miller CA, et al. Mortality of coronary artery bypass grafting before and after the advent of angioplasty. Ann Thorac Surg 1989;47:493–7.[Abstract]
8. Naunheim KS, Fiore AC, Wadley JJ, et al. The changing mortality of myocardial revascularization: coronary artery bypass and angioplasty. Ann Thorac Surg 1988;46:666–74.[Abstract]
9. Naunheim KS, Fiore AC, Wadley JJ, et al. The changing profile of the patient undergoing coronary artery bypass surgery. J Am Coll Cardiol 1988;11:494–8.[Abstract]
10. Leeman DE, McCabe CH, Faxon DP, et al. Use of percutaneous transluminal coronary angioplasty and bypass surgery despite improved medical therapy for unstable angina pectoris. Am J Cardiol 1988;61:38G–44G.[Medline]
11. Foster ED, Fisher LD, Kaiser GC, et al. Potential for percutaneous transluminal coronary angioplasty before initial and repeat coronary artery bypass grafting in the Coronary Artery Surgery Study (CASS) Registry population. Am J Cardiol 1984;53:112C–5C.[Medline]
12. Hochberg MS, Gielchinsky I, Parsonnet V, Hussain SM, Mirsky E, Fisch D. Coronary angioplasty versus coronary bypass. Three-year follow-up of a matched series of 250 patients. J Thorac Cardiovasc Surg 1989;97:496–503.[Abstract]
13. Jang GC, Block PC, Cowley MJ, et al. Relative cost of coronary angioplasty and bypass surgery in a one vessel disease model. Am J Cardiol 1984;53:52C–5C.[Medline]

This article has been cited by other articles:

				P. Massoudy, M. Thielmann, N. Lehmann, A. Marr, G. Kleikamp, A. Maleszka, A. Zittermann, R. Korfer, M. Radu, A. Krian, et al. Impact of prior percutaneous coronary intervention on the outcome of coronary artery bypass surgery: A multicenter analysis J. Thorac. Cardiovasc. Surg., April 1, 2009; 137(4): 840 - 845. [Abstract] [Full Text] [PDF]

				M. Thielmann, M. Neuhauser, S. Knipp, E. Kottenberg-Assenmacher, A. Marr, N. Pizanis, M. Hartmann, M. Kamler, P. Massoudy, and H. Jakob Prognostic impact of previous percutaneous coronary intervention in patients with diabetes mellitus and triple-vessel disease undergoing coronary artery bypass surgery J. Thorac. Cardiovasc. Surg., August 1, 2007; 134(2): 470 - 476. [Abstract] [Full Text] [PDF]

				M. Thielmann, R. Leyh, P. Massoudy, M. Neuhauser, I. Aleksic, M. Kamler, U. Herold, J. Piotrowski, and H. Jakob Prognostic Significance of Multiple Previous Percutaneous Coronary Interventions in Patients Undergoing Elective Coronary Artery Bypass Surgery Circulation, July 4, 2006; 114(1_suppl): I-441 - I-447. [Abstract] [Full Text] [PDF]

				J. M.D. van den Brule, L. Noyez, and F. W.A. Verheugt Risk of coronary surgery for hospital and early morbidity and mortality after initially successful percutaneous intervention Interactive CardioVascular and Thoracic Surgery, April 1, 2005; 4(2): 96 - 100. [Abstract] [Full Text] [PDF]

				H. Kamiya, T. Ushijima, K. Mukai, C. Ikeda, K. Ueyama, and G. Watanabe Late patency of the left internal thoracic artery graft in patients with and without previous successful percutaneous transluminal coronary angioplasty Interactive CardioVascular and Thoracic Surgery, March 1, 2004; 3(1): 110 - 113. [Abstract] [Full Text] [PDF]

				M.S. Barakate, J.M. Hemli, C.F. Hughes, P.G. Bannon, and M.D. Horton Coronary artery bypass grafting (CABG) after initially successful percutaneous transluminal coronary angioplasty (PTCA): a review of 17 years experience Eur. J. Cardiothorac. Surg., February 1, 2003; 23(2): 179 - 186. [Abstract] [Full Text] [PDF]

				R. G. Johnson, C. Sirois, R. L. Thurer, F. W. Sellke, W. E. Cohn, R. E. Kuntz, and R. M. Weintraub Predictors of CABG Within One Year of Successful PTCA: A Retrospective, Case-Control Study Ann. Thorac. Surg., July 1, 1997; 64(1): 3 - 7. [Abstract] [Full Text]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Johnson, R. G. Articles by Weintraub, R. M. Search for Related Content PubMed PubMed Citation Articles by Johnson, R. G. Articles by Weintraub, R. M.