Therapy of Severe Rheumatic CarditisComparison of Adrenocortical Steroids and Aspirin

By GABOR CZONICZER, M.D., FRANCISCO AMEZCUA, M.D.,SALVATORE PELARGONIO, M.D., AND BENEDICT F. MASSELL, M.D.

ALTHOUGH adrenocortical steroid therapyhas been available for 14 years, its value

in acute rheumatic fever still remains contro-versial. The Cooperative Study 1 failed to dem-onstrate any significant difference between as-

pirin and steroids with regard to their effect on

residual heart disease. On the other hand, ex-

tensive data collected at the House of theGood Samaritan 2 have shown that steroids inlarge dosage cause a considerably higher inci-dence of regression and disappearance of sig-nificant murmurs than does aspirin. The studyof Dorfman et al.,3 which evaluates the effectof prolonged therapy with aspirin, steroids,and the combination of both drugs, comes tothe same conclusion.Our clinical experience for some time has

shown that the difference in results with ster-oids and aspirin therapy is especially strikingin very ill patients with rheumatic carditis andthat in such patients steroid therapy may belifesaving. Because most clinicians have hadthe same experience, aspirin has, since the ad-vent of steroids, rarely been used in verysevere rheumatic carditis. This fact has hin-dered the collection of reliable data comparingthe effect of aspirin with that of steroids insevere rheumatic carditis and explains whysuch data cannot be found in the medical liter-ature.

In an attempt to overcome this obstacle we

have reviewed our experience with very illpatients observed at the House of the Good

From the House of the Good Samaritan, Children'sHospital Medical Center, and the Department of Pedi-atrics, Harvard Medical School, Boston, Massachusetts.

Supported by Grant HE 04957 from the NationalHeart Institute, U. S. Public Health Service, Grant511-F from the Massachusetts Heart Association, andby a training grant (for Dr. Amezcua) from the Barn-stable County Chapter of the Massachusetts HeartAssociation.

Circulation, Volume XXIX, June 1964

Table 1

Clinical Material: 137 Cases of Actute RheumaticFever with Congestive Heart Failure *

CHF without CHF with TotalTherapy pericarditis pericarditis cases

No therapy 28 14 42Aspirin 27 15 42Steroids 37 16 53Total 92 45 137

* Not included in this table are eight additionalcases of pericarditis without congestive heart failure.

Samaritan during the past two and a half dec-ades. The purpose of this study is to presentthe results of our critical review.

Clinical Material and MethodsTable 1 shows our clinical material, which is

comprised of ward patients observed at the Houseof the Good Samaritan from 1939 through 1962.We included in this study only patients who hadsevere rheumatic carditis as indicated by conges-tive heart failure or pericarditis.As this table shows, we evaluated 137 cases of

congestive heart failure due to active rheumaticcarditis. These 137 patients include 42 who weregiven no antirheumatic therapy, 42 who weretreated with aspirin, and 53 who were treated withsteroids. The next to the last column of table 1shows that some of the patients also had pericar-ditis in addition to congestive heart failure. Not in-cluded in this table are eight other patients whohad pericarditis without congestive heart failure.Thus, the total number of cases evaluated is 145.Our schedule of treatment was as follows: The

patients in the aspirin group generally received 40mg. of aspirin per pound of body weight per dayfor 12 weeks, but some had a somewhat shortercourse. The treatment for the patients who wereincluded in the steroid group consisted in the greatmajority of cases of a 12-week course of prednisoneor dexamethasone, the dose being 60 mg. and 6mg., respectively, per day for the first 3 weeks;this dose was tapered gradually during the next 9weeks. The patients who were labeled as receiv-ing `no therapy" were given no antirheumatic

813

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

CZONICZER ET AL.

Table 2

Characteristics of 137 Cases of Acute Rheumatic Fever with Congestive Heart Failure

Therapy

No therapy

Aspirin

Steroids

umkLerof Age of

cases patients

42 2-17avg. -9.6

42 4-17avg. 10.3

53 2-17avg. 8.2

SeverityofCHF

2.0

Averagedurationof CHFprior toonset oftherapy oradmission,

wk.

8.8

Initial attack ofrheumatic feverCases Per cent

24 57

2.0 6.4 28 67

2.4 5.7 44 83

drugs at all or a minimal amount of aspirin, a dosethat could not be considered as effective.

Table 2 shows the comparability of the 137 con-

gestive failure cases with regard to several featuresthat possibly might have influenced the course oftheir disease. Such features, listed in table 2, areas follows: age of the patient, severity and dura-tion of congestive heart failure, additional penicil-lin treatment, and whether the patient was treatedduring a first attack or during a recurrence. Theseverity of congestive heart failure was graded ona scale of 1 to 4. The average grade of severitywas about the same for each of the three groups,and there is no significant difference in the timeinterval that elapsed between the onset of conges-tive heart failure and the beginning of treatmentin the aspirin and in the steroid group. There is noappreciable difference in the average age of thepatients either, since only patients under 17 yearsof age were included in all three groups. However,the groups differ with regard to the proportion ofpatients observed in their initial attack of rheu-matic fever and the proportion treated with peni-cillin in addition to an antirheumatic drug. It isevident that there are relatively more first attacksand more penicillin-treated patients in the steroidgroup, a fact that makes comparison somewhatdifficult.

ResultsTo evaluate the results we have taken into

consideration the behavior of the congestiveheart failure-that is, whether it subsided or

worsened-and the death rate in the aspirinand in the steroid group within the first year

after onset of therapy.Table 3A shows the results of the different

therapeutic approaches. The proportion ofcases in which congestive heart failure sub-sided was only 36 per cent in the nontreated,

43 per cent in the aspirin-treated, but 92 percent in the steroid-treated group. Almost half(47 per cent) of the patients who were notgiven antirheumatic therapy died during hos-pitalization. In contrast, there were only twodeaths (4 per cent) in the group of 53 patientstreated with steroids, and of these two deaths,only one was actually due to rheumatic fever;the other was caused by complicating hemor-rhagic chickenpox. In the group of 42 patientstreated with aspirin there were nine deaths-adeath rate of 21 per cent. Although this rate isdefinitely less than that for the nontreatedgroup, it is well above the 4-per cent rate forthe steroid-treated patients. Statistical analysisshows that the differences for this table arehighly significant (p<O.OOOl). Thus it is evi-dent from this table that steroid therapy wasdefinitely superior to aspirin therapy in this se-ries of patients with severe rheumatic carditis.

Because we were especially interested in the

Table 3A

Results of Therapy in 137 Cases of Acute Rheu-matic Fever with Congestive Heart Failure

Worsenedor

Number appearedTherapy of case3 Subsided unchanged Died

No therapy 42 15 7 20(36%) (17%) (47%)

Aspirin 42 18 15 9(43%) (36%) (21%)

Steroids 53 49 2 2 *(92%) (4%) (4%)

* One death due to chickenpox.

Circulauion, Volume XXIX, June 1964

Additionalpenicillingiven

Cases Per cent

3 7

23 55

43 81

84

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

THERAPY OF SEVERE RHEUMATIC CARDITIS

of Acute Rheu-Heart Failure

NumberTherapy of cases Subsided Worsened Died

Aspirin 42 18 15 9(43%) (36%) (21%)

Steroids 53 49 2 2(92%) (4%) (4%)

Total 95 67 17 11

difference in the effect of aspirin versus ster-

oids, we analyzed the figures for these two

drugs alone, excluding the group that was

given no therapy (table 3B). A chi-square test

showed that the differences here also are

highly significant (p<O.Ol).Table 2 shows that there were more penicil-

lin-treated cases in the steroid group than inthe aspirin group. In order to obviate the effectof penicillin on results, both the aspirin andthe steroid groups were divided into two sub-groups: one subgroup comprised patientsgiven penicillin in addition to aspirin or ster-oids (table 4); the other subgroup was madeup of patients who were not given penicillin(table 5).Tables 4 and 5 show that the steroid-treated

patients responded definitely better than didthe aspirin-treated patients, whether or notadditional penicillin was given. Of the patientsreceiving penicillin (table 4), the percentageof cases that improved was only 48 in theaspirin group, but 88 in the steroid group; thedeath rate was 17 per cent in the aspirin group

and only 5 per cent in the steroid group. Thedifference is also evident in the patients whowere not given penicillin (table 5); in this sub-group there was 100-per cent improvement inthe steroid-treated patients and only 37-percent improvement in the aspirin-treated pa-

tients. The differences are highly significantfor both tables 4 and 5 (p<O.O1).A second possible factor in addition to peni-

cillin treatment, which was considered in eval-uating results, is whether the patients were

treated during a first attack or during a recur-

rence of rheumatic fever. The proportion ofCirculation, Volume XXIX, June 1964

first attacks was slightly greater in the steroidgroup than in the aspirin group. Therefore, wecompared the results of aspirin and steroidtreatment separately in patients with first at-tacks (table 6) and in patients with recur-rences (table 7).

It is obvious that here too the steroid-treated patients did better than those treatedwith aspirin. When only first attacks wereanalyzed (table 6), the percentage of patientswho improved was only 47 in the aspirin groupin contrast to 91 in the steroid-treated group(p<O.Ol). In the table including only thosepatients who were treated during a recurrence(table 7) the difference between the effect ofaspirin and that of steroids is not statisticallysignificant if we consider the three grades ofeffect (i.e., "subsided,"' worsened,' and "died")separately. However, if we combine the groupof patients whose congestive heart failureworsened with the group of deaths (last twocolumns) and compare this new group withthe group of patients whose congestive heartfailure subsided during treatment, the differ-ence in effect of aspirin versus steroids be-comes significant (p<O.O5).The results presented in tables 4 to 7 show

Table 4

Results of Therapy in 66 Patients with SevereRheumatic Carditis to Whom, in Addition to Anti-rheumatic Therapy, Penicillin Was Given

NumberTherapy of cases Subsided Worsened Died

Aspirin 23 11 8 4(48%) (35%) (17%)

Steroids 43 38 3 2(88%) (7%) (5%)

Table 5

Results of Therapy in 29 Patients with SevereRheumatic Carditis to Whom No Penicillin WasGiven

NumberTherapy of cases Subsided Worsened Died

Aspirin 19 7 7 5(37%) (37%) (26%)

Steroids 10 10 0 0(100%)

Table 3B

Results of Therapy in 95 Casesmatic Fever with CongestiveTreated with Aspirin or Steroids

&15

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

CZONICZER ET AL.

Table 6

Results of Therapy in 72 Patients with First At-tacks of Acute Rheumatic Fever and CongestiveHeart Failure

NumberTlherapy of cases Subsided Worsene(d I)ied

Aspirin 28 13 9 6(47%) (32%) (21%)

Steroids 44 40 2 2(91%) (4.5%) (4.5%)

Table 8

Results of Therapy in 36 Patients, All Treatedfrom 1956 Through 1962

Therapy

Aspirin

Steroids

Numnberof cases Subsided

16 9( 56%)

20 18(90%)

Worsened Died

5(31%)2

(10%)

2(13%)

0

clearly that the superior effect of steroids overaspirin, demonstrated by this study, cannot beattributed to the fact that there were morepenicillin-treated patients and slightly morefirst attacks in the steroid group than in theaspirin group.There is another factor, however, that should

be taken into consideration when we are com-paring the effect of the two drugs. We arereferring to the theory that the pattern ofacute rheumatic fever is changing and thatthis disease has decreased in severity duringthe last four decades.4' To exxclude the in-fluence of this factor we evaluated separatelythe cases which were treated in the time pe-riod between January 1953 and December 1962.These 36 patients, of whom 16 were given as-pirin and 20 were given steroids, wvere quitecomparable in every aspect.Table 8 shows that there was a 90-per cent

improvement in the steroid group in contrastto a 56-per cent improvement in the aspiringroup. There were two deaths in the aspiringroup but no deaths in the steroid group. Al-though the number of cases in Table 8 is small,statistical evaluation of these cases shows thatthe p value for this figture is 0.05. This means

Table 7

Results of Therapy in 23 Patients with Recturrenceof Rheumatic Fever with Severe RheuimaticCarditis

Nu1X-iberTherapy of cases Subsided Worsened Died

Aspirin 14 5 6 3(36%) (43%) (g1%)

Steroids 9 8 1 0(89%) (11%)

that the difference between the effect of as-pirin and that of steroids is probably significanteven when we include in o-ur evaluation onlycases treated in the same period of time. Itseems reasonable to infer that the presumedchanging pattern of acute rheumatic fever can-not be a determining factor in the superior re-sult of steroids in this series of severe rheu-matic carditis.

Still another factor that may have influencedour results is the appearance of recurrent acuterheumatic fever during the course of therapy.It is well known that recurrences have a se-rious effect on the prognosis of valvular disease.Hence, if the distribution of recurrences wereto be uneven in the three groups, this factshould be taken into consideration when eval-uating the effect of therapy. To be able tomake this evaluation, we reviewed all the casesin our study regarding recurrences. It turnedout that there were definitely more recurrencesin the so-called "no therapy" group than in theother two groups. This fact could be partlyresponsible for the high death rate in the "notherapy" group. However, we have found onlyone patient who had a recurrence of rheumaticfever during aspirin treatment and who died.If we excluded this one patient from the study,the percentage of deaths wouild drop soslightly (from '21 per cent to 19 per cent) thatit wouild not influence the evaluation of ourresults. Hence, the inferior effect of aspirincannot be attributed to the influience of rheu-matic fever recurrences.Table 9 shows the results of steroid and as-

pirin therapy in pericarditis, especially theeffects on the pericardial friction rub. The lastcolumn of this table shows that there were 11deaths in the total group of 53 patients with

(ircltaione. Vol/u-me XXIX, June 1964

816

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

THERAPY OF SEVERE RHEUMATIC CARDITIS

Table 9

Pericarditis

AppearedTotal during

Therapy cases therapy Died

No therapy 17 4 6Aspirin 16 3 4Steroids 20 0 1Total 53 7 11

pericarditis. Six of these deaths occurred in 17patients given no antirheumatic drugs, andfour of the deaths occurred in 16 patientstreated with aspirin. In the group of 20 pa-tients with pericarditis who were treated withsteroids, there was only one death.As table 9 shows, in not one instance did we

observe the appearance of a pericardial fric-tion rub during steroid treatment. However, afriction rub appeared during observation infour of 17 patients who did not receive anyantirheumatic drugs and in three of 16 patientsduring aspirin administration. From this ex-perience we are inclined to infer that peri-carditis-a severe manifestation of rheumaticfever-is more effectively suppressed by ster-oids than by aspirin.

DiscussionAs we have mentioned in the introduction,

the relative value of aspirin and steroids in thetreatment of acute rheumatic fever is contro-versial. Some of the studies-such as the Co-operative Study 1 and the Combined Rheu-matic Fever Study "-failed to demonstrate anysignificant difference in effect between the twodrugs when the disappearance of murmurswas used as a measure of efficiency. On theother hand, data collected in the House of theGood Samaritan 2 and in the study of Dorfmanet al.3 have shown a considerably higher inci-dence of regression and disappearance of sig-nificant murmurs in steroid-treated patientsthan in aspirin-treated patients.

All the above-mentioned studies have beenconcerned with the entire scale of acute rheu-matic fever cases, from very mild to verysevere. However, we have noticed for sometime that the superior effect of steroids is es-pecially striking in cases of severe rheumaticCirculation, Volume XXIX, June 1964

carditis with congestive heart failure and thatthis drug, unlike aspirin, may be lifesaving insuch cases. Nevertheless, we could not finddata in the medical literature comparing theeffect of these two drugs exclusively in verysevere rheumatic carditis, i.e., in cases in whichthe difference in effectiveness may be the mostevident.The studies that encompass the entire scale

of acute rheumatic fever are not suitable forevaluation of the effectiveness of aspirin andsteroids in patients with severe rheumatic(arditis. Thus in the Cooperative Study 1 thedistribution of cases with congestive heart fail-tire among the three groups was uneven; tothe aspirin group-evidently by chance-therewere alloted relatively fewer patients withcongestive heart failure (3.8 per cent) than tothe ACTH and cortisone groups (12.5 per centand 10 per cent, respectively).The study of Markowitz and Kuttner 7 shows

clearly the excellent effect of cortisone insevere rheumatic carditis. This study, however,has the defect that it is limited to cases treatedwith steroids and does not include a compara-tive group treated with aspirin. Accordingly,the conclusion of the authors of this study isthe statement that "further studies based onlarge series of cases with adequate controlsare needed to establish the value of this (corti-sone) regimen."Moreover, in the Combined Rheumatic

Fever Study,6 set up especially for the purposeof finding out whether the more potent drugin the treatment of acute rheumatic fever isaspirin or steroids, all five patients with severecongestive heart failure who were included inthis study were treated with prednisone. Twoof these patients were originally in the steroidgroup; however, the medication of the threeremaining patients, who originally were allotedto the aspirin group, was changed to steroidsbecause "the chief investigator decided thatacute symptoms of these critically ill patientsmight be controlled more effectively withprednisone than with aspirin." It is evident thatunder such circumstances this study also can-not be used to evaluate which of the two drugsis more effective in severe rheumatic carditis.

817

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

CZONICZER ET AL.

This example shows clearly that the com-parison of the effect of steroids versus aspirinin severe rheumatic carditis has been hinderedby the fact that most investigators think as wedo-namely, that it is not suitable to treat withaspirin a patient who is in congestive heartfailure due to acute rheumatic carditis. This isthe reason why, since the advent of steroids, sofew patients with severe rheumatic carditishave been treated with aspirin and why ran-dom selection of drugs in severely ill patientshas been generally thought to be contraindi-cated.To offset these circumstances we had to at-

tempt to compare cases that were not ran-domly selected. In our study, whereas all thesteroid-treated cases were observed after 1949,approximately half of the aspirin-treated pa-tients were observed prior to that year. Thisdifference in the time period in which our pa-tients were treated makes it understandablethat the characteristics of our groups are notsimilar; there are relatively more penicillin-treated patients and relatively more first at-tacks in the steroid group than in the aspiringroup.

This difficulty could be overcome only byregrouping our cases and by building smallergroups which were comparable. As tables 4 to7 show, it turned out that the superiority ofsteroids in the management of severe carditisis so evident that the difference between theeffect of aspirin and that of steroids proved tobe significant even if we compared these new,smaller groups of patients.The most difficult task was to eliminate the

influence of the supposed changing pattern ofacute rheumatic fever. If the supposition isvalid that acute rheumatic fever is a less se-vere disease today than it was twvo or threedecades ago, the cases treated with steroids 20to 25 years ago are not comparable to the casestreated with steroids during the last 10 years.We tried to overcome this difficulty, too, bycomparing exclusively the course of 12 aspirin-treated patients and 16 steroid-treated pa-tients, all of whom were observed during thesame period, namely, within the last 7 years,and whose characteristics wvere all similar.

Since the effect of steroids was significantlybetter than that of aspirin, even in these pa-tients, we can state that the supposed chang-ing pattern of acute rheumatic fever did notenter into our results in this group of patients.Thus, when the supposed influence of the timefactor can be excluded, steroids would seemdefinitely superior to aspirin in the treatmentof severe rheumatic carditis.Another reason why in the studies men-

tioned above the superiority of steroids couldnot be demonstrated may be that the usualmethod for evaluating the effect of an anti-rheumatic drug cannot be used in cases withsevere carditis. In all of these studies evalua-tion was made by taking into considerationthe percentage of patients who lose their mur-murs within 1 to 5 years after onset of therapy.This method, which is effective in cases withmoderately severe carditis, is not feasible inpatients with congestive heart failure. This istrue because the murmur indicating rheumaticvalvulitis disappears in only a small percent-age of all patients who also had congestiveheart failure, whether they were given aspirinor steroids.2 Thus the superior effect of ster-oids cannot be demonstrated if we use the dis-appearance of murmurs as an indicator.

In our study, therefore, it was not the dis-appearance of significant murmurs that wasused for evaluation but rather the behavior ofcongestive heart failure and the death ratewithin the first year after onset of treatmentwith aspirin and steroids, respectively.

ConcIusionsThe data presented show that in our series

of 145 patients with severe rheumatic carditissteroids had a definitely better therapeutic ef-fect than aspirin. This conclusion is evidentfrom the markedly lower death rate and fromthe faster and higher recovery rate in our ster-oid-treated group and also from the fact thatnot one of our patients developed a pericardialfriction rub while on steroids. Appearance orworsening of congestive heart failure occurredalso much more rarely during steroid therapythan during aspirin treatment.There are many variables that may have in-

Circulatizon, Volume XXIX, June 1964

818

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

THERAPY OF SEVERE RHEUMATIC CARDITIS

fluenced these excellent results: additionalpenicillin therapy, the supposed changing pat-tern of acute rheumatic fever during the lastdecades, recurrence of rheumatic fever duringtreatment, and the question of whether we aredealing with a first attack or with a recurrence.Although the distribution of our patients re-garding these factors was uneven in the as-pirin and in the steroid group, we were able toshow (through regrouping of our cases) thatthe superior effect of steroids cannot be attrib-uted to uneven distribution.

In spite of the fact that treatment in this se-ries could not be determined by random selec-tion, the observations collected during thisstudy strongly support our clinical experiencethat steroids in large dosage are the treatmentof choice in severe rheumatic carditis withcongestive heart failure or pericarditis.

AcknowledgmentThe authors are most grateful to Dr. Hugo Muench

for assistance in the statistical analysis of the data pre-sented in this study.

References1. The evolution of rheumatic heart disease in chil-

dren. Five-year report of a cooperative clinicaltrial of ACTH, cortisone, and aspirin. Circula-tion 22: 503, 1960.

2. MASSELL, B. F., JHAVERI, S., CZONICzER, G., ANDBARNET, R.: Treatment of rheumatic fever andrheumatic carditis: Observations providing abasis for the selection of aspirin or adrenocorti-cal steroids. M. Clin. North America 45: 1349,1961.

3. DORFMAN, A., GRoss, J. I., AND LORINCZ, A. E.:The treatment of acute rheumatic fever. Pedi-atrics 27: 692, 1961.

4. BLAND, E. F.: Declining severity of rheumaticfever: a comparative study of the past four dec-ades. New England J. Med. 262: 597, 1960.

5. Editorial: Acute rheumatic fever, a changing dis-ease. J.A.M.A. 182: 1035, 1962.

6. Combined Rheumatic Fever Study Group: A com-parison of the effect of prednisone and acetyl-salicylic acid on the incidence of residual rheu-matic heart disease. New England J. Med. 262:895, 1960.

7. MARKOWITZ, M., AND KUTTNER, A. G.: The effectof intensive and prolonged therapy with corti-sone and hydrocortisone in first attacks of rheu-matic carditis. Pediatrics 16: 325, 1955.

Reports of Medical Cases, with Reference to Morbid AnatomyPreface by Richard Bright-1827

The morbid appearances which present themselves on the examination ofthose who have died with dropsical effusion, either into the large cavities ofthe body or into the cellular membrance, are exceedingly various: and it oftenbecomes a matter of doubt how far these organic changes are to be regardedas originally causing or subsequently aiding the production of the effusion, andhow far they are to be considered merely as the consequence either of theeffusion or of some more general unhealthy state of the system. If it were possi-ble to arrive at a perfect solution of these questions, we might hope to obtainthe highest reward which can repay our labours,-an increased knowledge ofthe nature of the disease, and improvement in the means of its treatment.-Original Papers of Richard Bright on Renal Disease. Edited by A. ARNOLD OSMAN.London, Oxford University Press, 1937, p. 1.

Circulation, Volume XXIX, June 1964

819

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from

and BENEDICT F. MASSELLGABOR CZONICZER, FRANCISCO AMEZCUA, SALVATORE PELARGONIO

and AspirinTherapy of Severe Rheumatic Carditis: Comparison of Adrenocortical Steroids

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1964 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.29.6.813

1964;29:813-819Circulation. 

http://circ.ahajournals.org/content/29/6/813.citationlocated on the World Wide Web at:

The online version of this article, along with updated information and services, is

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. and Rights Question and Answer

Permissionsthe Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle column ofClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:

by guest on May 11, 2018

http://circ.ahajournals.org/D

ownloaded from