Anticoagulants, antiaggregants or nothing following carotid endarterectomy?

Eur J Vasc Surg 7, 364-369 (1993)

Anticoagulants, Antiaggregants or Nothing Following Carotid Endarterectomy?* Georg Bischof, Thomas Pratschner, Martin Kail, Martina Mittlb6ck, Edwin Turkof, Stefan Puig, Peter Polterauer and Georg Kretschmer University Clinic of Surgery L Vienna, Austria Carotid endarterectomy (TEA) has proven to be beneficial for symptomatic patients. Anticoagulation (AC) and antiplatelet therapy (ASA) have been shown to prolong life following vascular surgery in patients with occlusive arterial disease (PAOD). To determine whether ASA or AC prolong life after TEA, retrospective analysis was undertaken, since cerebral haemorrhage is associated with the use of both drugs, especially AC. Between 1979-1986, 328 patients with stenotic lesions of the carotid bifurcation were operated upon electively. Patient survival and causes of death were the primary end points of the analysis. Recent data were obtainedfrom the Austrian Central Bureau of Statistics. Cumulative survival rates were calculated by KaplanMeier estimation and differences determined by Breslow and Mantel tests. 36 patients were on AC, 157 on ASA and 135 remained without medication (O-group). Since the common risk factors in PAOD were unevenly distributed between groups,~ stepwise Cox regression model was applied which revealed age (p < 0.01), cardiac pathology (p < 0.01) and diabetes (p < 0.05) as relevant for survival. Therefore, ASA patients and O-group patients were selected and matched, employing the aforementioned prognostic criteria, and compared to the patients on long-term AC for various indications (vein bypass surgery, myocardial infarction, pulmonary embolism; i.e. data-matching). The median postoperative survival was 7.72 years for ASA and 8.48 years for AC, compared to 6.07 years for the O-group (p = 0.0095 Breslow, p = 0.477 Mantel). There was no significant difference between AC and ASA treated patients. Irrespective of medication, the causes of death were well balanced, and no higher incidence of intracerebral haemorrhage was detected. Our analysis demonstrates that in the long-term, antiaggregants and anticoagulants seem to be equally effective after TEA, and that postoperative pharmacotherapy is an important factor for patient survival, probably by reducing postoperative coronary events. Key Words: Carotid endarterectomy; TEA; Postoperative survival; Antiaggregants; Anticoagulants; Matched-pair analysis.

Introduction Atheromatous narrowing of the carotid bifurcation is the most common cause of large-vessel atherothrombotic stroke. Such lesions may lead to bruits, minor strokes or transient ocular as well as cerebral ischaemic attacks in the territory of the carotid artery, or, less often, to severe, disabling or even fatal strokes.1 Carotid endarterectomy (TEA) combined with postoperative pharmacotherapy versus medical therapy alone reduces the risk of severe stroke or death among symptomatic patients with high-grade carotid stenosis. This has been recently shown in two major trials conducted independently from each other (NASCET, ECST). 2'3 In cerebrovascular disease the value of antiplatelet drugs (ASA) and oral anticoagulants (AC) has been assessed in two meta-analyses and did not

produce convincing evidence in favour of medication. 4"5 Nevertheless, ASA reduced the incidence of vascular events in general, as demonstrated in the Oxford trial meta-analysis. 6 The importance of surgical pharmacotherapy in peripheral arterial disease has also been addressed, 7 and the "Pros" and "Cons" of ASA and AC have been discussed recently, s Following TEA a higher risk of cerebral haemorrhage due to antithrombotic drugs has been claimed, s In order to investigate which drug is best to use in endarterectomised patients, a retrospective evaluation, employing the technique of matched-pair analysis, was undertaken. The patients' death was used as the primary end point of the analysis.

Patients and Methods * Presented at the 6th Annual Meeting of the European Society for Vascular Surgery, Athens, September 1992.

Between 1979-1986, 328 patients with a stenosis of Please address all correspondence to: Dr G. Bischof, University the carotid bifurcation, either in the asymptomatic Clinic of SurgeryI, Alser Strasse 4, A-1090Vienna, Austria. clinical state (n = 108) or after one or more transient 0950-821X/93/070364+06 $08.00/0© 1993Grune & StrattonLtd.

Drug Therapy After Carotid Endarterectomy

Table 1. Causes of anticoagulant treatment before carotid TEA

Peripheral arterial occlusive disease

n = 24

Myocardial infarction, ischaemia, arrhythmia

n=3

Vascular surgery

n=7

Aortocoronary bypass

n=2

ischaemic attacks (n = 220), were eligible for analysis. Patients were o p e r a t e d u p o n electively and a standard carotid e n d a r t e r e c t o m y was carried out, generally w i t h o u t intraluminal shunting. The arterio t o m y was closed b y direct suture. Thirty-six patients w e r e receiving long-term anticoagulant treatment for various indications (Table 1) before surgery and w e r e continued o n p h e n p r o c o u m o n tablets (3 m g each) thereafter. A T h r o m b o t e s t ( " N y c o m e d " , I m m u n o AG, Vienna, Austria) bet w e e n 5 - 1 2 % was c o n s i d e r e d safe and within therapeutic range. 157 patients received u p to 1.5 g acetylsalicylic acid (ASA) daily, and 135 patients r e m a i n e d w i t h o u t antithrombotic medication (0-group). The decision to prescribe AC, ASA or n o t h i n g was the surgeon's. Patients were followed up by regular appointm e n t s o n an out-patient basis at 3-month intervals initially a n d twice a year thereafter. Particular attention was paid to patient compliance. Primary e n d points of evaluation w e r e patient survival and causes of death. Follow-up r a n g e d from 0.02-17 years. The m e d i a n length of follow-up was 6.8 years (interquantile range 5-9).

365

information concerning survival status of patients was obtained from the Austrian Central Bureau of Statistics (Osterreichisches Statistisches Zentralamt, Vienna, Austria) and transferred to the c o m p u t e r at least once a year (in Austria current legislation warrants reporting of residency and of causes of death, as well as obtaining permission for p o s t - m o r t e m examination.* Analysis of data revealed that different risk factors of P A O D were u n e v e n l y distributed a m o n g patient groups. Therefore, a Cox regression analysis 13 was u s e d to explore these factors, w h i c h indep e n d e n t l y influenced the o u t c o m e in a most relevant way. The factors were t h e n e m p l o y e d for a 2:1 case: control ratio for post data matching. 14 Matching refers to the pairing of one or m o r e controls to each case on the basis of "similarity" with respect to the selected variables. Variables to be considered for m a t c h i n g are those, that w e r e explored as risk factors b y a regression analysis. One attempts to have a constant case:control ratio (1:1, 1:2, 1:more), d e p e n d i n g on the size of the control series available, in order to g u a r a n t e e matching of the entire case series. O n e s h o u l d expect to screen four to five times as m a n y controls to match the case series. W e have used a 1:2 ratio, that was f o u n d feasible according to the n u m b e r of patients eligible. Survival data were expressed as m e d i a n survival, a n d asymptotic standard errors w e r e given w h e r e available.

Results

Statistical methods

The d o c u m e n t a t i o n system of the Austrian Society for Vascular Surgery was used to enter details of all surgical t r e a t m e n t p r o c e d u r e s in a c o m p u t e r (180 variables per surgical intervention, more than 8000 operations since 1965). We had access to the IBM 4381 main frame c o m p u t e r at the Faculty of Medicine. Data were stored on-line and retrieved with Statistical Analysis System (SAS) software. The use of SAS and the Biomedical Dixon Program (BMDP-1L Life) facilitated survival analysis and regular outpatient a p p o i n t m e n t s , as described before. 9 The first carotid TEA of each patient was considered for analysis. Cumulative survival rates were calculated b y Kaplan-Meier estimation 1° and possible differences b e t w e e n groups d e t e r m i n e d with Breslow 11 and Mantel tests 12 in a bitailed manner. The most recent

Data of 328 patients, o p e r a t e d u p o n for stenotic lesions of the carotid bifurcation, were available for evaluation. Patients received postoperative pharmac o t h e r a p y or were left u n t r e a t e d until d e a t h or until time of final evaluation b y the e n d of 1991. The usual risk factors of peripheral arterial disease w e r e distribu t e d a m o n g g r o u p s as s h o w n in Table 2. During "1. Under the Austrian Act on Reporting of Residency (Meldegesetz, BGBL 1972/336) as amended, residents et aI. (e.g. hotels for their guests) have to report their place of residence (as all changes thereof) to the local authorities (local maior) or to the police (§1 Meldegesetz). Failure to comply with such reporting obligation may result in administrative fines of up to AS 3000, or up to 2 weeks in prison (§16 Meldegesetz). 2. Under §27 subpara. 4 (BGBL 1983/60, Personenstandsgesetz) as amended the following applies: if a person dies in a hospital, then the director of the clinic or the physician who performed the coroner's inquest has to report to the local authorities (Personenstandsbeh6rde) the cause of death to enable the local authorities to pass that information on to the Austrian Central Office of Statistics (Osterr. Statistisches Zentralamt). Eur J Vasc Surg Vol 7, July 1993

366

G. Bischof et aL

Table 2. Distribution of risk factors between groups before matching

Variable

Description

AC n = 36

Age

~<55 55-65 >65

6 10 20

20 47 90

16 50 69

Diabetic state

non-diabetic diabetes I (oral medication) diabetes II (insulin-dependent)

24 10 2

103 42 10

73 44 18

Cardiac pathology

nil arrhythmia ischaemia/infarction other

11 5 14 6

63 11 58 25

56 10 53 16

Hyperlipidaemia

yes no

24 12

95 62

97 38

Sex

male female

28 8

111 46

95 40

Hypertension

<160/90 mmHg > 160/90 mmHg

25 11

97 15

88 47

Smoking habits

non-smoker <10 cigarettes/day >10 cigarettes/day

10 6 18

46 15 81

39 16 74

Clinical stage

I (asymptomatic)

14

II (TIA)

22

39 118

55 80

f o l l o w - u p 19 p a t i e n t s d i e d in t h e A C - g r o u p (53%), 56 in the A S A - g r o u p (36%) a n d 84 in the 0 - g r o u p (62%). T h e difference in t h e p r o b a b i l i t y of survival w a s significant (p = 0.0233 Mantel, p = 0.0001 Breslow) (Fig. 1), b u t w a s of limited relevance. 100

-,

35

90 80 70 60 50 40

3

so

12

20 10 I_~

0

12

I

3

I

r

4

I

5

p

I

I

6

r

7

I

I

8

I

I

I

I

9 10

r

I

I

11 12

Years

Fig. 1. Kaplan-Meier estimation (unmatched data) of cumulative survival of all patients (n = 328) after TEA. 135 patients had no therapy, 157 patients received ASA and 36 patients AC therapy. Test statistics therapy v s . no therapy: p = 0.0233 Mantel, p = 0.0001 Breslow. Standard error did not exceed 10% in any group. Numbers above or below curves indicate patients at risk [numbers at risk at the time of last death (median survival): no therapy (--), 13 at 8.8 years (5.77 + 1.15 years); ASA (A), 11 at 7.7 years (6.84 + 0.21 years); AC (A), 7 at 8.6 years (8.48 + 0.59 years)]. Eur J Vasc Surg Vol 7, July 1993

ASS n = 157

0 n - 135

Employing the Cox regression model a variety of factors i n f l u e n c e d the probability of survival, e x p l a i n e d as p-value a n d relative risk ( s u m m a r i s e d in Table 3). A g e , i s c h a e m i c cardiac d i s e a s e a n d diabetic m e t a b o l i c state a d v e r s e l y i n f l u e n c e d survival. B o t h postoperative medical therapies improved the outc o m e in a significant w a y . U s i n g the t h r e e a f o r e m e n t i o n e d factors t h e t e c h n i q u e of d a t a - m a t c h i n g (1:2, c a s e : c o n t r o l ratio) w a s a p p l i e d to a d j u s t the i n c i d e n c e of i m p o r t a n t p r o g n o s t i c f a c t o r s . T h u s , t h e A S A - a n d t h e 0 - g r o u p e a c h c o n t a i n e d twice as m a n y p a t i e n t s t h a n t h e A C - g r o u p , a n d the risk factors w e r e r e a s o n a b l y b a l a n c e d a m o n g g r o u p s (Table 4). T h e m e d i a n p o s t o p e r a t i v e survival time w a s 8.48 (S.E. 0.59) y e a r s in the A C - g r o u p a n d 7.72 y e a r s in t h e A S A - g r o u p , w h e r e a s in t h e u n t r e a t e d g r o u p t h e m e d i a n s u r v i v a l w a s 6.07 (S.E. 0.97) years. T h e differe n c e b e t w e e n s u r v i v a l c u r v e s w a s statistically signific a n t (p = 0.0095 Breslow, p = 0.0477 Mantel) (Fig. 2). N o significant difference b e t w e e n t r e a t m e n t g r o u p s w a s n o t i c e d (p = 0.61 Breslow, p = 0.85 Mantel). A t the time of e v a l u a t i o n , 17 p a t i e n t s w e r e alive in the A C - g r o u p a n d all of t h e m w e r e o n A C treatm e n t , w h i c h c o u l d easily be c o n t r o l l e d b y c h e c k i n g t h e p a t i e n t ' s A C chart. T h e c h a r t is carried b y t h e p a t i e n t a n d s h o w s dates a n d v a l u e s of all t h r o m b o t e s t control measurements. T h e l e a d i n g c a u s e s of d e a t h w e r e i s c h a e m i c m y o -

Drug Therapy After Carotid Endarterectomy

oo\ 80

Table 3. Stepwise Cox regression analysis to explore significant prognostic factors; end point: patient survival Variable

p

Risk ratio

Age

0.0001

1.78

Sex

0.899

1.02

ASS

0.031

0.65

AC

0.017

0.50

Cardiac pathology (ischaemia)

0.0001

2.16

Cardiac pathology (arrhythmia)

0.098

1.78

Other cardiac pathology

0.083

1.65

Hypertension

0.445

0.87

Diabetes I (oral medication)

0.003

2.25

Diabetes II (insulin-dependent)

0.246

1.26

Hyperlipidaemia

0.069

1.40

Smoking (<10 cigarettes/day)

0.857

1.05

Smoking (>10 cigarettes/day)

0.673

0.92

Clinical stage

0.292

0.83

367

7o~ - ~

~

~ 9

5o

40 E

3

30 2O 10 L

0

I

1

i

I

2

I

I

3

I

I

4

I

I

5

r

I

I

l

6 7 Years

I

I

8

I

I

9

I

P

10

I

I

=

11 12

Fig. 2. Kaplan-Meier estimation (matched data) of cumulative survival after data-matching of 36 patients with AC v s . 72 patients with ASA vs. 72 patients without postoperative medication (1:2). Test statistics therapy v s . no therapy: p = 0.0477 Mantel, p = 0.0095 Breslow. Standard error did not exceed 10% in any group. Numbers above or below curves indicate patients at risk ]numbers at risk at the time of the last death (median survival): no therapy (--), 10 at 8.8 years (6.07 + 0.97 years); ASA (A), 9 at 7.7 years (7.72 years); AC (&), 7 at 8.6 years (8.48 + 0.59 years)]. No standard error could be expressed for the median postoperative survival of the ASA-group because the survival rate did not fall below 45.6%.

Table 4. Risk factors for data-matching (1:2, i.e. each AC patient was matched with 2 ASA patients and 2 patients without therapy) AC n = 36

ASA n = 72

0 n = 72

6 10 20

12 20 40

12 20 40

non-diabetic diabetes I (oral medication) diabetes II (insulindependent)

24 10

47 19

52 16

2

6

4

Cardiac pathology

nil arrhythmia ischaemia/infarction other

11 5 14 6

24 8 30 10

27 8 28 9

Hyperlipidaemia

yes no

24 12

45 27

59 13

Sex

male female

28 8

57 15

47 25

Hypertension

< 160/90 mmHg >160/90 mmHg

25 11

39 33

48 24

Smoking habits

non-smoker <10 cigarettes/day >10 cigarettes/day

10 6 18

17 4 42

16 11 43

Clinical stage

I (asymptomatic) II (TIA)

14 22

22 50

30 42

Variable Age

Description ~<55 55-65

>65 Diabetic state

Eur J Vasc Surg Vol 7, July 1993

368

G. Bischof e t al.

Table 5. Causes of death in the matched patient groups AC n = 19

ASA n = 25

0-group n = 46

Ischaemic myocardiopathy

68%

40%

43%

Cerebrovascular insufficiency

10%

12%

11%

Pulmonary cause

10%

24%

13%

Other (malignancy, urogenital disease, sepsis, etc.)

12%

24%

33%

(n = number of dead patients; ASA- and 0-group contain twice as many patients as the AC-group, therefore percentages are given.)

cardiopathy (68% AC, 40% ASA, 39% 0-group) and cerebral deaths which were, however, equally distributed among groups (10% AC, 12% ASA, 11% 0group; see also Table 5).

Discussion

The results of an univariate analysis followed by a Cox regression model showed that postoperative pharmacotherapy prolonged patient survival after carotid endarterectomy. In a matched pair analysis, using a control group of sufficient size to render close matching possible, the results were reproduced. This study dealt with a limited series of patients from a single centre, but with the advantages of uniform assessment, management and complete follow-up, facilitated by information obtained from the Austrian Central Bureau of Statistics. Patients were followed up for a median of 6.8 years. The Kaplan-Meier estimates showed a constant decline over the years, pointing out the fact that the risk of dying from underlying cardiovascular disease remained fairly constant. It has been recognised that carotid artery disease is associated with coronary artery disease, and that cardiac events determine the fate of these patients. We have previously shown that AC prolongs survival following infrainguinal vein bypass surgery 15 and that antiaggregants prolong postoperative survival time after TEA. 16 The TEA trial produced evidence that this improvement might be caused by lowering the incidence of cardiac events in the medicated group, which is in accordance with the results of the Oxford trial meta-analysis. 6 Both drugs, and AC in particular, seem to have the potential danger of inducing cerebral haemorrhage. In our study the incidence of cerebral deaths was around 10% and distributed equally among all groups, implying that the risk of haemorrhage might have been overestimated in the past. This impression Eur J Vasc Surg Vol 7, July 1993

is reinforced by the results of the Dutch anticoagulant trial. 17 The mode of action of platelet function stabilising agents, on the one hand, and the fact that arterial thrombosis is platelet mediated, on the other, are strong arguments for the prescription of ASA. Furthermore, ASA is relatively safe, but the optimal dose and formulation are still open to debate. Gastrointestinal side effects, which are obviously dose related, also have to be considered. The risk of AC therapy largely depends upon the availability of a system to control the level of anticoagulation as frequently as necessary in any individual patient. The patient's compliance is easy to check for AC, whereas major difficulties have been encountered in Aspirin-trials and discordances of results depending upon the kind of analysis--"intention to treat" or "on t r e a t m e n t " - have been recently recognised. 18 The results of our retrospective analysis confirm the efficacy of anticoagulants as well as antiaggregants to prolong survival after carotid endarterectorny. The risk of fatal cerebral events was not prohibitive. In the absence of any prospective trial, it seems justified to prescribe long-term postoperative antithrombotic pharmacotherapy of either kind.

Acknowledgement We are indebted to the Austrian Central Office of Statistics for providing us with the survival data of the patients.

References 1 KISTLERJP,BUONANNOFS, GRESSDR. Carotid endarterectomy-specific therapy based on pathophysiology. N Engl J Med 1991; 325: 505-507. 2 NORTH AMERICAN SYMPTOMATIC CAROTID ENDARTERECTOMY TRIAL (NASCET) STEERINGCOMMITTEE. North American Symptomatic Carotid Endarterectomy Trial. Methods, patient characteristics and progress. Stroke 1991; 22: 711-720.

Drug Therapy After Carotid Endarterectomy

3 EUROPEANCAROTIDSURGERYTRIALISTS'COLLABORATIVEGROUP.

4

5 6

7 8 9

10 11

MRC European Carotid Surgery Trial: interim results for symptomatic patients with severe (70-99%) or with mild (0-29%) carotid stenosis. Lancet 1991; 337: 1235-1243. SZE PC, REITMAN D, PINCUS MM, SACKS HS, CHALMERSTC. Antiplatelet agents in the secondary prevention of stroke: metaanalysis of the randomized control trials. Stroke 1988; 19: 436442. JONAS S. Anticoagulant treatment in cerebrovascular disease: review and meta-analysis. Stroke 1988; 19: 1043-1048. ANTIPLATELETTRIALISTS' COLLABORATION. Secondary prevention of vascular disease by prolonged antiplatelet treatment. BMJ 1988; 296: 320-331. DORMANDYJ. Surgical pharmacotherapy. Eur J Vasc Surg 1989; 3: 379-380. VAN URK H, KRETSCHMERG. What is the role of oral anticoagulants and platelet inhibition in perhipheral vascular surgery? Eur J Vasc Surg 1990; 4: 553-555. PRATSCHNERTH, KRETSCHMERG, PRAGERM, et al. Antiplatelet therapy following carotid bifurcation endarterectomy. Evaluation of a controlled clinical trial. Prognostic significance of histologic plaque examination on behalf of survival. Eur J Vasc Surg 1990; 4: 1-5. KAPLANEL, MEIER P. Non-parametric estimation from incomplete observations. J Am Statist Assoc 1958; 53: 457-481. BRESLowN. A generalized Kruskal-Wallis test for comparison of

12 13 14 15

16

17 18

369

k-samples subject to unequal patterns of censorship. Biometrika 1970; 57: 579-594. MANTELN. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966; 50: 163-170. Cox DR. Regression models and life-tables (with discussion). J R Statist Soc 1972; B 34: 187-220. SCHLESSELMANJJ. Case control studies. Design, conduct, analysis. In: Monographs in Epidemiology and Biostatistics. Oxford University Press, 1982; 105-123. KRETSCHMERG, WENZLE, SCHEMPERM, et al. Influence of postoperative anticoagulant treatment on patient survival after femoro-popliteal vein bypass surgery. Lancet 1988; April 9: 797799. KRETSCHMERG, PRATSCHNERTH, PRAGERM, et al. Antiplatelet treatment prolongs survival after carotid bifurcation endarterectomy. Analysis of the clinical series followed by a controlled trial. Ann Surg 1990; 211: 317-322. DE SMIT P, VAN URK H. Dutch oral anticoagulation trial. Acta Chir A 1992; 1: 5-7. FRANKSPJ, SIAN M, KENCHINGTONGF, ALEXANDERCE, POWELL JT, EEMOROPOPLITEALBYPASSTRIALPARTICIPANTS.Aspirin usage and its influence on femoro-popliteal vein graft patency. Eur ] Vasc Surg 1992; 6: 185-188.

Accepted 4 January 1993

Eur J Vasc Surg Vol 7, July 1993