- Email: [email protected]
Recanalization of Femoropopliteal Occlusive Lesions: A Comparison of Long-term Clinical, Color Duplex US, and Arteriographic Follow-up
Recanalization of Femoropopliteal Occlusive Lesions: A Comparison of Long-term Clinical, Color Duplex US, and Arteriographic Follow-up1 Dammis Vroegindeweij, MD Alexander V. Tielbeek, MD Jacob Buth, MD, PhD Marjolijn J. van Kints, MD Guido H. M. Landman, MD, PhD Willem P. T. M. Mali, MD, PhD Index terms: Arteries, femoral, 93.1282 Arteries~popliteal~93.1282 Arteries, stenosis or obstruction, 93.721 Arteries, transluminal angioplasty, 93.1282
JVIR 1995;6:331-337 Abbreviations: ABI = ankle-brachial index, PSV = peak systolic velocity, SE = standard SVS'SCVS = Society for Vascular Surgery/International Society for Cardiovascular Surgery
PURPOSE: To assess the merits of clinical examination, color-flow duplex ultrasound (US),and arteriography in the follow-up of patients who have undergone femoropopliteal artery recanalization for occlusive disease. PATIENTS AND METHODS: Recanalization of the occluded femoropopliteal artery was attempted in 62 patients. Follow-up included clinical examination, ankle-brachial blood pressure measurement, and duplex US scanning at 4-month intervals during the first year, at 6-month intervals during the second year, and once a year thereafter. Failure of recanalization included substantial restenosis or reocclusion of the treated segment. Arteriography was performed at the end of the first year or earlier if recurrence was suspected. Agreement of clinical findings with those of duplex US and those of arteriography was determined with K statistics; a K value of greater than 0.75 represented excellent agreement. RESULTS: Recanalization was technically successful in 51 patients (82%).Clinical patency was 63%(standard error [SE], 6%)after 1 year, 56%(SE, 7%)after 2 years, and 46%(SE, Wo) after 3 years. When technical failures were included, the patency rate at duplex US was 58%(SE, 6%)after 1year, 40%(SE, 7%)after 2 years, and 33%(SE, 8%)after 3 years. The patency rate at arteriography was 53%(SE, 7%)after 1year, 33%(SE, 7%)after 2 years, and 30%(SE, 8%)after 3 years. When arteriographic examination was considered the standard of reference, diagnostic accuracy in the identification of recurrent lesions was 94%at duplex US (K = 0.88) and 74%at clinical examination (K = 0.51). CONCLUSION: Rates of restenosis or occlusion detected at followup with duplex US and arteriography were comparable. However, clinical examination alone helped detect fewer cases of recurrent disease.
ITis difficult to analyze the differ-
From the Departments of Radiology (D.V.,A.V.T., M.J.V.K., G.H.M.L.) and Vascular surgery (J.B.), catharina ~ ~ tal, Michelangelolaan 2,5623 EJ Eindhoven,TheNetherlands, andtheDepa*ment of Radiology (W.P.T.M.),University State Hospital, Utrecht. Received August 15,1994; revision requested October 3; revision received November 14; accepted NOvember 21. Address reprint requests to D.V.
o SCVIR,
1995
ent results of endovascular intervention in lower extremity arteries that have been reported in the literature; ~ ~ i there are great differences in the severity of treated lesions, and a wide variety of methods and for failure have been used in follow-up assessment (1-19). When results of follow-up were evaluated based only on worsening of the symptoms and decrease in peripheral pulses, as
was the case in earlier studies, patency rates as high as 70% at 5 years were reported (16). However, restenosis or reocclusion is not always indicated by the recurrence of symptoms or the absence of distal pulses because collaterals may develop or the pattern of walking activities may change in an elderly patient. The use of an ankle-brachial blood pressure index (ABI) at followup has been viewed as an important
331
332
Journal of Vascular and Interventional Radiology
May-June 1995
step toward greater objectivity in the assessment of the results of treatment. Although ABI has been able to help identify complete recurrent occlusion, it has shown low sensitivity in the diagnosis of stenosis (20,211. Furthermore, a standardized classification of findings that combines clinical symptoms and ABI results, as suggested in the guidelines of the Society for Vascular Surgerydnternational Society for Cardiovascular Surgery (SVSdSCVS), has rarely been used in reports of endovascular intervention (22). The use of arteriography at follow-up has mainly been limited to cases in which symptoms have returned or ABI has deteriorated, conditions most often associated with complete occlusion (2,7-9,11,15-17). Unfortunately, occlusions are often beyond the stage a t which recanalization can be performed successfully, and even if they can be recanalized, there is little chance of durable secondary patency. Identification of developing stenotic lesions and intervention a t follow-up may help prevent progression to complete reocclusion. Colorflow duplex ultrasound (US) is an accurate, noninvasive method that can be used serially to help detect development of restenosis a t the site of intervention and occurrence of new disease (20,23-27). Until recently, no studies to our knowledge have compared clinical findings (with the standardized SVSdSCVS criteria), duplex US findings, and arteriography data (22). There have been several studies of endovascular treatment, but they did not stratify the results according to whether the lesions were femoropopliteal stenoses or complete occlusions; the patency differs considerably between these two categories (8,9,28,29). Although the probability of recurrence of occlusion is greater in patients with a primary occlusion, the usefulness of color duplex US a t follow-up may also be greater in these patients. Therefore, we used color duplex US to examine all patients who had undergone recanalization and
balloon angioplasty of a complete fernoropopliteal occlusion. Our aim was to evaluate and compare the long-term clinical, duplex US, and arteriography findings in patients who had undergone recanalization of a fernoropopliteal artery occlusion.
PATIENTS AND METHODS -
Study Group From January 1988 to July 1993, prospective follow-up studies were performed in 62 patients with angiographically proved fernoropopliteal artery occlusion. The mean duration of follow-up was 23 months (range, 0-69 months). There were 45 men (73%)and 17 women (27%),with an average age of 64 years (range, 4778 years). The risk factors for atherosclerosis seen in these patients were hypertension in five patients (8%),diabetes mellitus in 16 (26%), history of smoking in 45 (72%),and coronary disease in nine (14%). Symptoms at presentation in all patients were classified according the SVSIISCVS guidelines (22). Fifty patients (81%)had mild to moderate intermittent claudication (class I [category 1-21), 11patients (18%) had severe claudication (class I [category 3]), and one patient (1%)had rest pain (class 11). The length of the occlusions ranged from 0.5 to 15 cm, with a mean length of 4.7 cm; 27 arterial segments were less than 4 cm long, 26 lesions were more than 4 cm and less than 7.5 cm long, and nine lesions were more than 7.5 cm long. The crural vessel runoff was good (two or three open crural vessels) in 48 patients. Coexistent second and third stenoses were observed in five patients. Only one limb per patient was treated. All patients gave consent to undergo regular duplex US studies and arteriography during follow-up. The following inclusion criteria were used: (a) lesions had to be confined to the fernoropopliteal arteries; (b) occlusions had to be suitable for recanalization, that is, no concomitant lesions beyond the primary occlusion
that would substantially compromise the iliac and femoral inflow or the popliteal and crural runoff; and (c) no previous endovascular or operative procedures could have been performed in the ipsilateral femoropopliteal area. Preintervention assessment included physical examination, determination of Doppler ABI, and aortography examination with runoff studies that included intraarterial digital subtraction angiography in the evaluation of the fernoropopliteal and crural arteries. Recanalization was performed with a guide wire in 44 patients, with high-frequency ablation in 16, and with a Rotacs catheter (Dr. P. Osypka GMBH, Grenzach-Wyhlen, Germany) in two. All patients then underwent balloon angioplasty. After the occlusion had been traversed by means of a guide wire or high-frequency ablation, a balloon catheter was introduced, and angioplasty was performed. During the procedure, 5,000 U of heparin was administered intraarterially. Arteriography was then performed in two orthogonal directions. When more sophisticated guide wires became available, we began to use only a guide wire to recanalize the occluded femoropopliteal artery. Postprocedural monitoring included peripheral pulse palpation and ABI measurements. Patients usually were discharged the second day after the procedure. All patients were given 80 mg of aspirin orally a t least 1day before the intervention, and this dosage was continued indefinitely.
Methods of Follow-up Follow-up included clinical examination and determination of the ABI 1week after the intervention. These procedures were combined with ABI measurement after graded treadmill exercise and duplex US scanning; all procedures were performed routinely at 4-month intervals during the first year, a t 6-month intervals during the second year, and once a year or when symptoms recurred thereafter. Arteriography was performed earlier only if there was
Vroegindeweij et a1
333
Volume 6 Number 3
clinical deterioration or if stenosis was suspected a t color duplex US examination; in all other patients, arteriography was performed 1year after the intervention. If the results of both the clinical and the duplex US examinations were negative, no further arteriography examinations were performed, and the recanalization site was considered patent. Color duplex US scanning was performed with a Quantum Angiodynograph (Philips Medical Systems, Shelton, Conn). During the examination the entire femoropopliteal segment was examined from the groin to the popliteal trifurcation. Stenotic segments were identified by means of locally increased velocities and poststenotic turbulence characterized by coded color changes. Changes in diameter as well as Doppler velocity signals were recorded for normal and diseased segments. To aid in differentiation between restenosis and new stenosis, duplex US findings were drawn on a diagram of the femoropopliteal arteries, with the superficial femoral artery divided into four segments (Sl, S2, S3, S4) and the popliteal arteries divided into two segments (Pl, P2). The degree of restenosis was assessed by means of the local increase in the peak systolic velocity (PSV) at the site of the stenosis, as compared with the PSV of a nearby normal arterial segment. The ratio of these two values was calculated; a PSV ratio of a t least 2.0 was considered to indicate a stenosis of 50% or more, according to the duplex US criteria for classification of peripheral arterial obstructions proposed by Jager et a1 (24). Reocclusion was characterized by means of the absence of detectable Doppler signals a t the treated segments and increased collateral flow. Definitions, End Points, and Statistical Analysis Clinical symptoms, ABI, and postexercise ankle blood pressure were classified according the SVSI ISCVS scale of postprocedural improvement (22). With this scale, -3
indicates deterioration, including increased symptoms and decreased ABI; 0 indicates no change in symptoms; +1 indicates either improvement of one symptom category or increase in ABI of more than 0.10; +2 indicates improvement of one symptom category along with an increase in ABI of more than 0.10; and +3 indicates marked improvement in symptoms along with an ABI of more than 0.90. An improvement of less than +1 was considered a failure of treatment. The procedure was defined as technically successful if, with visual estimation, the completion arterioeram demonstrated successful recanalization and residual stenosis of less than 30% in two orthogonal projections. Morphologic success or patency was assessed by identifying recurrent disease at the site of initial intervention; the patency rate was calculated se~aratelvwhen there was recurrent disease and new lesions in other segments. In addition, recurrent disease was subdivided into com~leteobstruction. defined as reocclusion, and incomplete obstruction. defined as restenosis. The Dresence of recurrent or new lesions was determined by means of duplex US and confirmed with arteriography. The arteriographic assessment of the degree of stenosis, in two orthogonal projections, was based on a ratio of the diameter of the stenotic site and that of a nearby normal site. The most severe stenosis was used for analysis. An arteriographically depicted diameter reduction of a t least 50% was considered a treatment failure. Correlation between clinical and duplex US findings versus those of arteriography, the standard of reference, was performed with two-by-two tables for calculation of accuracy of agreement, sensitivity, and specificity. The K statistic was used to determine the agreement between tests (30). The maximum value of K is 1, which represents perfect agreement. A K value of 0 represents agreement that occurs purely by chance. A K value of greater than 0.75 represents excelu
lent agreement; a value between 0.4 and 0.75 represents fair to good agreement; and a value of less than 0.4 represents poor agreement. This comparative analysis included only values that represented duplex US observations made at the time of corresponding arteriography. In addition to clinical deterioration, a postexercise ABI of less than 0.90 was correlated with arteriography and duplex US. Patency rates were statistically determined with the Kaplan-Meier life-table method. A diameter reduction seen a t arteriography of a t least 50% and a PSV ratio of a t least 2.0 at duplex US were considered to represent morphologic failure of the intervention. Results a t follow-up in the treatment group were calculated according to the "intention-to-treat" principle; that is, early technical failure and attempted but abandoned procedures were considered end points of patency. The Fisher exact test was used to compare discrete variables between groups. Continuous variables were compared with the Mann-Whitney test, and for paired comparisons the Wilcoxon test was used. RESULTS Technical Results and Complications The occluded vessel could not be recanalized in nine patients (14%). Failure was due to extensive dissection with unsuccessful attempts to reenter the vessel lumen in five patients. Vessel wall perforation was the reason for termination of the procedure in an additional four patients. There were two patients with residual stenosis and 40% reduction in diameter. Technically successful recanalization was achieved in 51 patients (82%).In three of nine patients in whom recanalization failed, further treatment was conservative, and these patients were without symptoms a t the end of follow-up. In five of these patients, early endarterectomy was required, and in the
334
Journal of Vascular and Interventional Radiology
May-June 1995
one remaining patient, a bypass procedure was performed electively. Small dissections were found in eight patients (13%).These lesions were not considered complications, as technical success (residual stenosis of no more than 30%)still could be achieved. No difference in ABI was observed in the treated leg as compared with the healthy contralateral leg. In addition, no occlusions were observed within the first month in these patients.
PATENCY
- CLINICAL
loo%
DUPLEX
.--.---- ARTERIO
..,
-. .-------.--.
16
Clinical Results at Follow-up During follow-up, clinical deterioration categorized as less than +1occurred in 20 patients. The cumulative rate of maintained clinical success categorized as at least +1 was 63% (standard error [SEI, 6%) after 1year, 56% (SE, 7%)after 2 years, and 46% (SE, 9%)after 3 years (Figure). In one patient in whom there was no clinical improvement immediately after successful recanalization, improvement occurred at follow-up. In two patients with residual stenosis of 40%, clinical improvement was maintained. The sensitivity of the clinical classificationwas 61% compared with arteriographically demonstrated diameter reduction of at least 50%; the specificity was 95%, and the accuracy was 74% (K= 0.51). The ABI before clinical deterioration was 0.89 + 0.11 compared with 0.78 0.17 thereafter. The postexercise ABI decreased significantly after clinical deterioration from 0.84 f 0.16 to 0.51 0.16. Sensitivity in the identification of restenosis by means of postexercise ABI (if the threshold is a postexercise ABI of less than 0.90) compared with arteriography was 81%. Specificity was 90%, and accuracy was good at 84%(K = 0.78).
+
+
Results of Color Duplex US at Follow-up Serial duplex US examinations during follow-up helped detect 11 reocclusions and 21 restenoses (Table). Most recurrences (15 of 21 restenoses, eight of 11reocclusions)
0%I 0
I-...
I
1
I
I
I
1
2
3
4
5
I
FOLLOW-UP IN YEARS
Figure. Cumulative maintained patency a t follow-up is summarized for findings a t clinical examination (failure is represented by SVS/ISCVS classification of c1) (I), at duplex US (failure defined as PSV ratio of a t least 2.01, and a t arteriography (failure defined as a t least 50% diamter reduction). Numbers represent patients a t risk a t different intervals.
occurred within the first year after intervention. Cumulative primary patency, defined as an absence of restenosis of at least 50% or reocclusion, was 58% (SE, 6%)after 1 year, 40% (SE, 7%) after 2 years, and 33% (SE, 8%)after 3 years (Figure). New stenoses occurred in segments other than the initially treated segments in nine patients. Six of these lesions developed in limbs that also had restenosis at the treated site, and three occurred in limbs that had a fully patent site of recanalization. When restenoses, reocclusions, and new stenoses combined were considered to represent treatment failure, the cumulative patency was 58% (SE, 7%)after 1 year, 39% (SE, 7%) after 2 years, and 31% (SE, 9%)after 3 years. The sensitivity of duplex US compared with arteriography was 91%, the specificity was loo%, and the accuracy was excellent at 94% (K = 0.88).
Results of Arteriography at Follow-up All patients underwent examination with arteriography. Reocclusion was found in 11 of the 51 patients in whom recanalization had been technically successful (Table). Restenosis of at least 50% had developed in 22 femoropopliteal arteries (Table). When restenosis and reocclusion combined were considered treatment failures, the cumulative patency rate at follow-up was 53% (SE, 7%) after 1year, 33% (SE, 7%)after 2 years, and 30% (SE, 8%)after 3 years (Figure). New stenoses of at least 50% occurred in 12 limbs (24%) during follow-up in segments other than those initially treated. Nine of these lesions developed in limbs in which there was a simultaneous restenosis, and three occurred in limbs with a patent recanalization site. When restenoses of at least 50%, reocclusions, and new stenoses of at least 50% were combined and were
Vroegindeweij et a1
335
Volume 6 Number 3
Summary of Lesions Detected by Means of Color-flowDuplex US and Arteriography
Lesion Restenosis of 250% Occlusion New stenosis of 250%
Color-flow Duplex US
Arteriography
21 11 9
22
considered to represent treatment failure, the cumulative patency a t follow-up was 53% (SE, 7%) after 1 year, 32% (SE, 7%) after 2 years, and 29% (SE, 8%) after 3 years. In one patient with a calcified femoral artery, duplex US demonstrated restenosis of less than 50%. and arteriography demonstrated stenosis of 65%. Three new stenoses were not depicted at duplex US because of severe calcification in the distal femoral artery a t the site of the Hunter canal.
DISCUSSION Several studies of the long-term results of recanalization of the femoropopliteal artery with either coldor hot-tip techniques have recently been reported (1-19). Despite the abundance of literature on this subject, it is difficult to find consistency in the outcomes because most study results rely on nonstandardized clinical data and incidental arteriography examinations a t follow-up (2,7-9,11,15-17). This approach to the analysis of long-term findings of stenosis is insufficient, as clinical examination is insensitive in the detection of restenosis and therefore is less suited to identify substantial differences between treatment modalities. The use of more sensitive tests such as duplex US would increase the likelihood of significant differences being found in the outcomes of treatment with recently developed and experimental techniques. Moreover, comparative clinical trials might require less time and smaller sample sizes and would be less expensive to accomplish. From a statistical point of view, a
11 12
more accurate follow-up method would help detect an end point of patency in more patients; this increases the power of the comparative analysis in a study population. For instance, in the present study the rate of clinical deterioration a t 3 years was 54% compared with the 67% failure rate determined with duplex US. Arteriography is currently used to help detect recurrence of stenosis or occlusion, clinical deterioration, and decrease in ABI. However, arteriography is an invasive and expensive method and cannot be performed serially at follow-up. And although arteriography provides precise anatomic data, it does not provide functional information about blood flow nor does it indicate flow abnormalities. Moreover, the actual time required for development of recurrent or new disease remains somewhat unknown, as arteriography cannot be performed repeatedly. Until recently, there have been no reports to our knowledge of the longterm clinical, arteriography, and duplex US results after recanalization with standardized criteria of failure. In the present study we used the SVSASCVS guidelines for classification of clinical results and the wellestablished criteria for duplex US of Jager et a1 (22,24). The 3-year clinical patency rate of 46% after recanalization found in our study was comparable to the results in other studies (8,9). Recanalization results assessed by means of arteriography and duplex US were lower than those determined clinically, with differences in 3-year patency rates of 13% and 16%, respectively. Thus clinical assessment is less sensitive in the detection of recurrent disease
than examination with arteriography and duplex US. These findings are in agreement with those of Miller et a1 (23), who reported that restenosis was depicted on color-flow images before it was found by means of ABI. Different threshold values of the PSV ratio chosen to represent recurrent disease correlate with different severity of disease. A PSV ratio of a t least 2.0 correlates well with the early development of restenosis, while a higher PSV ratio, of 2.5 for instance, may better correlate with the recurrence of clinical symptoms (31). The results of the present study, as well as those of other studies, suggest that the recurrence of symptoms combined with a decrease in resting ABI is not a sensitive indicator of restenosis (20,211. One factor may be that the development of collateral vessels abolishes the return of symptoms in cases of recurrent disease. It is our experience that for a complete clinical evaluation, the postexercise ABI should be included, as it seems to help identify more recurrent lesions than does resting ABI alone. In the present study the decrease in ABI, from 0.89 0.11 to 0.78 f 0.17, after clinical deterioration was rather modest compared with the significant decrease, from 0.84 f 0.16 to 0.51 f 0.16, in postexercise ABI. Compared with duplex US, however, the postexercise ABI is still less sensitive in the detection of recurrent disease (91% vs 81%). Nevertheless, postexercise ABI still appears a useful method in the identification of restenosis after intervention when color duplex US is not available. A sensitive, noninvasive method is needed to achieve a detailed study of the development of restenosis and reocclusion. This technique should enable determination of the location of a lesion and allow for differentiation between recurrent and new disease; the latter is usually caused by progression of atherosclerosis a t areas other than the treatment site. Color duplex US meets these requirements and is ideally suited to
+
Journal of Vascular and Interventional Radiology
336
-
May-June 1995
assessment of the durability of patency after the treatment of lesions in the femoropopliteal system. Findings in our study of 91% sensitivity and 100% specificity compared with arteriography confirm its reliability. The comparable accuracy of duplex US and arteriography in the diagnosis and grading of femoropopliteal lesions has been documented previously (20,23-27). In addition, duplex US can help differentiate between restenosis and reocclusion, and the severity of stenoses can be gauged and monitored with respect to their progression. We believe that regular follow-up with duplex US in patients who have undergone femoropopliteal recanalization is useful. Balloon angioplasty of stenotic lesions is simpler and has been associated with a higher procedural success rate than recanalization of occlusions (9). Moreover. the best long-term results with endovascular procedures in femoro~o~liteal disease have been obtained in patients with stenotic rather than occlusive disease (8,9, 16,19). Johnston et a1 (9) reported a success rate of 62% after 3 years in patients with stenoses, but that rate was only 46% in patients with occlusions. Also, Capek et a1 (8) and Gallino et a1 (19) reported long-term patency rates that were higher for stenoses than for complete occlusions (8,19). In our study, duplex US depicted a restenosis (21) considerably more often than a reocclusion (11)during follow-up. The importance of differentiating between restenosis and reocclusion was recently emphasized by Whyman et al (32), who said that the expected long-term outcome of reintervention was better for stenoses than for occlusions and that high-grade stenoses often progress into complete occlusions. Previous studies of the natural history of the development of femoropopliteal disease after endovascular procedures have demonstrated that restenoses occur much more frequently than progression of disease at the nontreated segments u
A
.
(33). New stenoses were detected in our study in nine patients by means of duplex US, and recurrent lesions were found in 32 patients. Differentiation of recurrent lesions in treated segments from new obstructions was achieved by arbitrarily dividing the superficial femoral artery into four segments and the popliteal artery into two segments. This scheme provided a way to record the location of the lesions demonstrated a t arteriography; a t follow-up with duplex US, the corresponding segments then could be identified with a ruler. This method of measurement in corresponding segments allowed us to compare disease identified by means of duplex US and arteriography and to use well-established criteria of severity. Because of the excellent agreement between duplex US and arteriographic findings, we now believe that reintervention in patients with high-grade, recurrent stenotic lesions may be planned without performance of arteriography for confirmation. This type of management of recurrent stenotic lesions will require less arteriography time, resulting in a more cost-effective use of health-care resources. We believe that the application of duplex US in the objective assessment of late results of percutaneous interventions should be expanded. This conclusion is in disagreement with the findings of Miller et a1 (231, who reported that none of their patients developed a late occlusion without an antecedent decrease in ABI; therefore, they stated their support for the use of ABI instead of the more expensive duplex US. Sensitivity with ABI was only 61% in our study, a finding which suggests that this method is not reliable in the detection of restenosis. However, we agree with Miller et a1 (23) that early detection of an increased PSV, which represents stenosis of at least 50%, a t duplex US without clinical complaints is not an indication for reintervention. However, during the follow-up period these stenoses could develop into high-grade stenoses or occlusions and could lead to clinical
deterioration. As stated previously, knowledge of impending occlusion with clinical complaints might allow for prevention in the preocclusive stage and might result in a better patency compared with occlusions (8,9,19). Our observations have strengthened our belief that use of a duplex US protocol after intervention is highly beneficial and may help improve overall patency. Especially during the first year of follow-up, we recommend a strict protocol, based on our observation that 23 of the 32 recurrences of stenosis and occlusion in this study were diagnosed in this period. After the first year, larger intervals between duplex US examinations are justified because of the lower incidence of recurrent lesions. In some cases in our study, there was no agreement between duplex US and arteriographic findings. Severe calcification of vessels often precludes visualization of stenotic lesions at duplex US because of the ultrasound beam attenuation. Calcification caused one restenosis and three new stenoses in Hunter canal to go undetected a t duplex US; these lesions were diagnosed only with arteriography. Despite these discrepancies, there was no statistically significant difference between the overall findings with arteriography and with duplex US. In conclusion, findings a t clinical examination, based on clinical symptoms and ABI, were insufficient in the detection of restenoses and reocclusions. Patency rates after 3 years of follow-up were comparable as determined with arteriography and duplex US. Duplex US has the advantage of being a noninvasive method that can be performed serially to help identify the development of restenosis. Timely identification by means of serial duplex US combined with reintervention may improve the overall results of recanalization of occlusions in the femoropopliteal arterial system. References 1. Berengoltz-Zlochin SN, Westerhof PW, Mali WPTM, et al. Nd:YAG la-
Vroegindeweij et al • 337 Volume 6 Number 3
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
ser-assisted angioplasty in femoropopliteal artery occlusions: "hot" versus "cold" recanalization with transparent contact probe. Radiology 1992; 182:409-414. Tobis JM, Conroy R, Stuart-Deutsch L, et al. Laser-assisted versus mechanical recanalization of the femoral arterial occlusions. Am J Cardiol 1991; 68:1079-1086. Luft CH, Horvath W, Oertl M, Haidinger D. Laser- versus rotationsangioplastie bei der rekanalisation chronischer femeropoplitealer arterien-verschliisse. Fortschr Rontgenstr 1993; 158:53-58. Odink HF, de Valois HC, Eikelboom BC. Femoropopliteal arterial occlusions: laser-assisted versus conventional percutaneous transluminal angioplasty. Radiology 1991; 181:61-66. Delcour CP, Wery D, Bank WO, Golzarian J, Struyven J. Arterial recanalization: hot-tip laser-assisted angioplasty versus hydrophilic guide-wire recanalization (abstr). Radiology 1991; 18l(P):266. Tobis J, Smolin M, Mallery J, et al. Laser-assisted thermal angioplasty in human peripheral artery occlusions: mechanism of recanalization. JAm Coll Cardiol1989; 13:15471554. Huppert PE, Duda SH, Helber U, Kansch KR, Clausen CD. Comparison of pulsed laser-assisted angioplasty in femoropopliteal artery occlusions. Radiology 1992; 184:363-367. Capek P, McLean GK, Berkowitz HD. Femoropopliteal angioplasty: factors influencing long-term success. Circulation 1991; 83(suppl 1):70-80. Johnston KW, Rae M, Hogg-Johnston SA, et al. Five-year results of a prospective study of percutaneous transluminal angioplasty. Ann Surg 1987; 206:403-413. Murray RR Jr, Hewes RC, White RI Jr, et al. Long-segment femoropopliteal stenoses: is angioplasty a boon or a bust? Radiology 1987; 162:473476. Hewes RC, White RI Jr, Murray RR, et al. Long-term results of superficial femoral artery angioplasty. AJR 1986; 146:1025-1029. Belli AM, Cumberland DC, Procter AE, Welsh L. Total peripheral ar-
13.
14.
15.
16.
17.
18. 19.
20.
21.
22.
23.
tery occlusions: conventional versus laser thermal recanalisation with a hybrid probe in percutaneous angioplasty-results of a randomized trial. Radiology 1991; 181:57-60. Zocholi G, Hotker U, Hake U, Storkel S, Schild H. Gefassrekanalisation mit einem thermo-kauter-katheter: in vitro-erfahrungen mit einem neuen kathetersysteem. Fortschr Rontgenstr 1988; 149:526528. Bonn J. Clinical utility oflaser recanalization in occluded peripheral arteries. Radiology 1991; 178:323325. Sanborn TA, Cumberland DC, Greenfield AJ, Welsh CL, Guben JR. Percutaneous laser thermal angioplasty: initial results and 1year follow-up in 129 femoropopliteal lesions. Radiology 1988; 168:121-125. Krepel VM, van Andel GJ, van Erp WFM, Breslau PJ. Percutaneous transluminal angioplasty of the femoropopliteal artery: initial and long-term results. Radiology 1985; 156:325-328. Martin EC, Fankuchen EI, Karlson KB, et al. Angioplasty for femoral artery occlusion: comparison with surgery. AJR 1981; 137:915-919. Gould DA. Thermal angioplasty using a high frequency ablator. J Intervent Radiol1990; 5:147-148. Gallino A, Mahler F, Probst P, Nachbur B. Percutaneous transluminal angioplasty of the arteries of the lower limbs: a 5 year follow-up. Circulation 1984; 70:619-623. Kinney EV, Bandyk DF, Mewissen MW, et al. Monitoring functional patency of percutaneous transluminal angioplasty. Arch Surg 1991; 126;743-747. Salles-Cunha SX, Andros G, Dulawa LB, Harris RW, Oblath RW. Changes in peripheral hemodynamics after percutaneous transluminal angioplasty. J Vasc Surg 1989; 10:338-342. Rutherford RB, Flanigan DP, Gupta SK, et al. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg 1986; 4:80-94. Miller BV, Sharp WJ, Shamma AR, Kresowik TF, Petrone S, Corson JD. Surveillance for recurrent stenosis after endovascular procedures. Arch Surg 1991; 126:867872.
24. Jager KA, Ricketts HJ, Strandness DE Jr. Duplex scanning for the evaluation of lower limb arterial disease. In: Bernstein EF, ed. Noninvasive diagnostic techniques in vascular disease. Philadelphia, Pa: Mosby-Year Book, 1985; 619-631. 25. Sachs D, Robinson ML, Marinelli DL, Perlmutter GS. Evaluation of the peripheral arteries with duplex US after angioplasty. Radiology 1990; 176:39-44. 26. Kohler TR, Nance DR, Cramer MM, et al. Duplex scanning for diagnosis of the aortoiliac and femoropopliteal disease: a prospective study. Circulation 1987; 76:1074-1080. 27. Polak JF, Karmel MI, Mannick JA, et al. Determination of the extent of lower-extremity peripheral arterial disease with color-assisted duplex sonography: comparison with angiography. AJR 1990; 155:10851089. 28. Adar R, Critchfield GC, Eddy DM. A confidence profile analysis of the results of femoropoliteal percutaneous transluminal angioplasty in the treatment of lower extremity ischemia. J Vasc Surg 1989; 10:57-67. 29. Jeans WD, Armstrong S, Cole SEA, Horrocks M, Baird RN. Fate of patients undergoing transluminal angioplasty for lower limb ischemia. Radiology 1990; 177:559-564. 30. Landis JR, Koch GG. The measurements of observer agreement for categorical data. Biometrics 1977; 33:159-174. 31. Legemate DA, Teeuwen C, Hoeneveld H, Ackerstaff RGA, Eikelboom BC. Spectral analysis criteria in duplex scanning of aortoiliac and femoropopliteal arterial disease. Ultrasound Med Biol1991; 17:769-776. 32. Whyman MR, Ruckley CV, Fowkes FGR. A prospective study of the natural history of femoropopliteal artery stenosis using duplex ultrasound. Eur J Vasc Surg 1993; 7:444-447. 33. Adams DH, Schoen FJ. Contemporary concepts in atherosclerosis pathology. In: White RA, ed. Atherosclerosis and arteriosclerosis: human pathology and experimental animal methods and models. Boca Raton, Fla: CRC, 1989; 49-86.
JVIR 1995;6:331-337 Abbreviations: ABI = ankle-brachial index, PSV = peak systolic velocity, SE = standard SVS'SCVS = Society for Vascular Surgery/International Society for Cardiovascular Surgery
PURPOSE: To assess the merits of clinical examination, color-flow duplex ultrasound (US),and arteriography in the follow-up of patients who have undergone femoropopliteal artery recanalization for occlusive disease. PATIENTS AND METHODS: Recanalization of the occluded femoropopliteal artery was attempted in 62 patients. Follow-up included clinical examination, ankle-brachial blood pressure measurement, and duplex US scanning at 4-month intervals during the first year, at 6-month intervals during the second year, and once a year thereafter. Failure of recanalization included substantial restenosis or reocclusion of the treated segment. Arteriography was performed at the end of the first year or earlier if recurrence was suspected. Agreement of clinical findings with those of duplex US and those of arteriography was determined with K statistics; a K value of greater than 0.75 represented excellent agreement. RESULTS: Recanalization was technically successful in 51 patients (82%).Clinical patency was 63%(standard error [SE], 6%)after 1 year, 56%(SE, 7%)after 2 years, and 46%(SE, Wo) after 3 years. When technical failures were included, the patency rate at duplex US was 58%(SE, 6%)after 1year, 40%(SE, 7%)after 2 years, and 33%(SE, 8%)after 3 years. The patency rate at arteriography was 53%(SE, 7%)after 1year, 33%(SE, 7%)after 2 years, and 30%(SE, 8%)after 3 years. When arteriographic examination was considered the standard of reference, diagnostic accuracy in the identification of recurrent lesions was 94%at duplex US (K = 0.88) and 74%at clinical examination (K = 0.51). CONCLUSION: Rates of restenosis or occlusion detected at followup with duplex US and arteriography were comparable. However, clinical examination alone helped detect fewer cases of recurrent disease.
ITis difficult to analyze the differ-
From the Departments of Radiology (D.V.,A.V.T., M.J.V.K., G.H.M.L.) and Vascular surgery (J.B.), catharina ~ ~ tal, Michelangelolaan 2,5623 EJ Eindhoven,TheNetherlands, andtheDepa*ment of Radiology (W.P.T.M.),University State Hospital, Utrecht. Received August 15,1994; revision requested October 3; revision received November 14; accepted NOvember 21. Address reprint requests to D.V.
o SCVIR,
1995
ent results of endovascular intervention in lower extremity arteries that have been reported in the literature; ~ ~ i there are great differences in the severity of treated lesions, and a wide variety of methods and for failure have been used in follow-up assessment (1-19). When results of follow-up were evaluated based only on worsening of the symptoms and decrease in peripheral pulses, as
was the case in earlier studies, patency rates as high as 70% at 5 years were reported (16). However, restenosis or reocclusion is not always indicated by the recurrence of symptoms or the absence of distal pulses because collaterals may develop or the pattern of walking activities may change in an elderly patient. The use of an ankle-brachial blood pressure index (ABI) at followup has been viewed as an important
331
332
Journal of Vascular and Interventional Radiology
May-June 1995
step toward greater objectivity in the assessment of the results of treatment. Although ABI has been able to help identify complete recurrent occlusion, it has shown low sensitivity in the diagnosis of stenosis (20,211. Furthermore, a standardized classification of findings that combines clinical symptoms and ABI results, as suggested in the guidelines of the Society for Vascular Surgerydnternational Society for Cardiovascular Surgery (SVSdSCVS), has rarely been used in reports of endovascular intervention (22). The use of arteriography at follow-up has mainly been limited to cases in which symptoms have returned or ABI has deteriorated, conditions most often associated with complete occlusion (2,7-9,11,15-17). Unfortunately, occlusions are often beyond the stage a t which recanalization can be performed successfully, and even if they can be recanalized, there is little chance of durable secondary patency. Identification of developing stenotic lesions and intervention a t follow-up may help prevent progression to complete reocclusion. Colorflow duplex ultrasound (US) is an accurate, noninvasive method that can be used serially to help detect development of restenosis a t the site of intervention and occurrence of new disease (20,23-27). Until recently, no studies to our knowledge have compared clinical findings (with the standardized SVSdSCVS criteria), duplex US findings, and arteriography data (22). There have been several studies of endovascular treatment, but they did not stratify the results according to whether the lesions were femoropopliteal stenoses or complete occlusions; the patency differs considerably between these two categories (8,9,28,29). Although the probability of recurrence of occlusion is greater in patients with a primary occlusion, the usefulness of color duplex US a t follow-up may also be greater in these patients. Therefore, we used color duplex US to examine all patients who had undergone recanalization and
balloon angioplasty of a complete fernoropopliteal occlusion. Our aim was to evaluate and compare the long-term clinical, duplex US, and arteriography findings in patients who had undergone recanalization of a fernoropopliteal artery occlusion.
PATIENTS AND METHODS -
Study Group From January 1988 to July 1993, prospective follow-up studies were performed in 62 patients with angiographically proved fernoropopliteal artery occlusion. The mean duration of follow-up was 23 months (range, 0-69 months). There were 45 men (73%)and 17 women (27%),with an average age of 64 years (range, 4778 years). The risk factors for atherosclerosis seen in these patients were hypertension in five patients (8%),diabetes mellitus in 16 (26%), history of smoking in 45 (72%),and coronary disease in nine (14%). Symptoms at presentation in all patients were classified according the SVSIISCVS guidelines (22). Fifty patients (81%)had mild to moderate intermittent claudication (class I [category 1-21), 11patients (18%) had severe claudication (class I [category 3]), and one patient (1%)had rest pain (class 11). The length of the occlusions ranged from 0.5 to 15 cm, with a mean length of 4.7 cm; 27 arterial segments were less than 4 cm long, 26 lesions were more than 4 cm and less than 7.5 cm long, and nine lesions were more than 7.5 cm long. The crural vessel runoff was good (two or three open crural vessels) in 48 patients. Coexistent second and third stenoses were observed in five patients. Only one limb per patient was treated. All patients gave consent to undergo regular duplex US studies and arteriography during follow-up. The following inclusion criteria were used: (a) lesions had to be confined to the fernoropopliteal arteries; (b) occlusions had to be suitable for recanalization, that is, no concomitant lesions beyond the primary occlusion
that would substantially compromise the iliac and femoral inflow or the popliteal and crural runoff; and (c) no previous endovascular or operative procedures could have been performed in the ipsilateral femoropopliteal area. Preintervention assessment included physical examination, determination of Doppler ABI, and aortography examination with runoff studies that included intraarterial digital subtraction angiography in the evaluation of the fernoropopliteal and crural arteries. Recanalization was performed with a guide wire in 44 patients, with high-frequency ablation in 16, and with a Rotacs catheter (Dr. P. Osypka GMBH, Grenzach-Wyhlen, Germany) in two. All patients then underwent balloon angioplasty. After the occlusion had been traversed by means of a guide wire or high-frequency ablation, a balloon catheter was introduced, and angioplasty was performed. During the procedure, 5,000 U of heparin was administered intraarterially. Arteriography was then performed in two orthogonal directions. When more sophisticated guide wires became available, we began to use only a guide wire to recanalize the occluded femoropopliteal artery. Postprocedural monitoring included peripheral pulse palpation and ABI measurements. Patients usually were discharged the second day after the procedure. All patients were given 80 mg of aspirin orally a t least 1day before the intervention, and this dosage was continued indefinitely.
Methods of Follow-up Follow-up included clinical examination and determination of the ABI 1week after the intervention. These procedures were combined with ABI measurement after graded treadmill exercise and duplex US scanning; all procedures were performed routinely at 4-month intervals during the first year, a t 6-month intervals during the second year, and once a year or when symptoms recurred thereafter. Arteriography was performed earlier only if there was
Vroegindeweij et a1
333
Volume 6 Number 3
clinical deterioration or if stenosis was suspected a t color duplex US examination; in all other patients, arteriography was performed 1year after the intervention. If the results of both the clinical and the duplex US examinations were negative, no further arteriography examinations were performed, and the recanalization site was considered patent. Color duplex US scanning was performed with a Quantum Angiodynograph (Philips Medical Systems, Shelton, Conn). During the examination the entire femoropopliteal segment was examined from the groin to the popliteal trifurcation. Stenotic segments were identified by means of locally increased velocities and poststenotic turbulence characterized by coded color changes. Changes in diameter as well as Doppler velocity signals were recorded for normal and diseased segments. To aid in differentiation between restenosis and new stenosis, duplex US findings were drawn on a diagram of the femoropopliteal arteries, with the superficial femoral artery divided into four segments (Sl, S2, S3, S4) and the popliteal arteries divided into two segments (Pl, P2). The degree of restenosis was assessed by means of the local increase in the peak systolic velocity (PSV) at the site of the stenosis, as compared with the PSV of a nearby normal arterial segment. The ratio of these two values was calculated; a PSV ratio of a t least 2.0 was considered to indicate a stenosis of 50% or more, according to the duplex US criteria for classification of peripheral arterial obstructions proposed by Jager et a1 (24). Reocclusion was characterized by means of the absence of detectable Doppler signals a t the treated segments and increased collateral flow. Definitions, End Points, and Statistical Analysis Clinical symptoms, ABI, and postexercise ankle blood pressure were classified according the SVSI ISCVS scale of postprocedural improvement (22). With this scale, -3
indicates deterioration, including increased symptoms and decreased ABI; 0 indicates no change in symptoms; +1 indicates either improvement of one symptom category or increase in ABI of more than 0.10; +2 indicates improvement of one symptom category along with an increase in ABI of more than 0.10; and +3 indicates marked improvement in symptoms along with an ABI of more than 0.90. An improvement of less than +1 was considered a failure of treatment. The procedure was defined as technically successful if, with visual estimation, the completion arterioeram demonstrated successful recanalization and residual stenosis of less than 30% in two orthogonal projections. Morphologic success or patency was assessed by identifying recurrent disease at the site of initial intervention; the patency rate was calculated se~aratelvwhen there was recurrent disease and new lesions in other segments. In addition, recurrent disease was subdivided into com~leteobstruction. defined as reocclusion, and incomplete obstruction. defined as restenosis. The Dresence of recurrent or new lesions was determined by means of duplex US and confirmed with arteriography. The arteriographic assessment of the degree of stenosis, in two orthogonal projections, was based on a ratio of the diameter of the stenotic site and that of a nearby normal site. The most severe stenosis was used for analysis. An arteriographically depicted diameter reduction of a t least 50% was considered a treatment failure. Correlation between clinical and duplex US findings versus those of arteriography, the standard of reference, was performed with two-by-two tables for calculation of accuracy of agreement, sensitivity, and specificity. The K statistic was used to determine the agreement between tests (30). The maximum value of K is 1, which represents perfect agreement. A K value of 0 represents agreement that occurs purely by chance. A K value of greater than 0.75 represents excelu
lent agreement; a value between 0.4 and 0.75 represents fair to good agreement; and a value of less than 0.4 represents poor agreement. This comparative analysis included only values that represented duplex US observations made at the time of corresponding arteriography. In addition to clinical deterioration, a postexercise ABI of less than 0.90 was correlated with arteriography and duplex US. Patency rates were statistically determined with the Kaplan-Meier life-table method. A diameter reduction seen a t arteriography of a t least 50% and a PSV ratio of a t least 2.0 at duplex US were considered to represent morphologic failure of the intervention. Results a t follow-up in the treatment group were calculated according to the "intention-to-treat" principle; that is, early technical failure and attempted but abandoned procedures were considered end points of patency. The Fisher exact test was used to compare discrete variables between groups. Continuous variables were compared with the Mann-Whitney test, and for paired comparisons the Wilcoxon test was used. RESULTS Technical Results and Complications The occluded vessel could not be recanalized in nine patients (14%). Failure was due to extensive dissection with unsuccessful attempts to reenter the vessel lumen in five patients. Vessel wall perforation was the reason for termination of the procedure in an additional four patients. There were two patients with residual stenosis and 40% reduction in diameter. Technically successful recanalization was achieved in 51 patients (82%).In three of nine patients in whom recanalization failed, further treatment was conservative, and these patients were without symptoms a t the end of follow-up. In five of these patients, early endarterectomy was required, and in the
334
Journal of Vascular and Interventional Radiology
May-June 1995
one remaining patient, a bypass procedure was performed electively. Small dissections were found in eight patients (13%).These lesions were not considered complications, as technical success (residual stenosis of no more than 30%)still could be achieved. No difference in ABI was observed in the treated leg as compared with the healthy contralateral leg. In addition, no occlusions were observed within the first month in these patients.
PATENCY
- CLINICAL
loo%
DUPLEX
.--.---- ARTERIO
..,
-. .-------.--.
16
Clinical Results at Follow-up During follow-up, clinical deterioration categorized as less than +1occurred in 20 patients. The cumulative rate of maintained clinical success categorized as at least +1 was 63% (standard error [SEI, 6%) after 1year, 56% (SE, 7%)after 2 years, and 46% (SE, 9%)after 3 years (Figure). In one patient in whom there was no clinical improvement immediately after successful recanalization, improvement occurred at follow-up. In two patients with residual stenosis of 40%, clinical improvement was maintained. The sensitivity of the clinical classificationwas 61% compared with arteriographically demonstrated diameter reduction of at least 50%; the specificity was 95%, and the accuracy was 74% (K= 0.51). The ABI before clinical deterioration was 0.89 + 0.11 compared with 0.78 0.17 thereafter. The postexercise ABI decreased significantly after clinical deterioration from 0.84 f 0.16 to 0.51 0.16. Sensitivity in the identification of restenosis by means of postexercise ABI (if the threshold is a postexercise ABI of less than 0.90) compared with arteriography was 81%. Specificity was 90%, and accuracy was good at 84%(K = 0.78).
+
+
Results of Color Duplex US at Follow-up Serial duplex US examinations during follow-up helped detect 11 reocclusions and 21 restenoses (Table). Most recurrences (15 of 21 restenoses, eight of 11reocclusions)
0%I 0
I-...
I
1
I
I
I
1
2
3
4
5
I
FOLLOW-UP IN YEARS
Figure. Cumulative maintained patency a t follow-up is summarized for findings a t clinical examination (failure is represented by SVS/ISCVS classification of c1) (I), at duplex US (failure defined as PSV ratio of a t least 2.01, and a t arteriography (failure defined as a t least 50% diamter reduction). Numbers represent patients a t risk a t different intervals.
occurred within the first year after intervention. Cumulative primary patency, defined as an absence of restenosis of at least 50% or reocclusion, was 58% (SE, 6%)after 1 year, 40% (SE, 7%) after 2 years, and 33% (SE, 8%)after 3 years (Figure). New stenoses occurred in segments other than the initially treated segments in nine patients. Six of these lesions developed in limbs that also had restenosis at the treated site, and three occurred in limbs that had a fully patent site of recanalization. When restenoses, reocclusions, and new stenoses combined were considered to represent treatment failure, the cumulative patency was 58% (SE, 7%)after 1 year, 39% (SE, 7%) after 2 years, and 31% (SE, 9%)after 3 years. The sensitivity of duplex US compared with arteriography was 91%, the specificity was loo%, and the accuracy was excellent at 94% (K = 0.88).
Results of Arteriography at Follow-up All patients underwent examination with arteriography. Reocclusion was found in 11 of the 51 patients in whom recanalization had been technically successful (Table). Restenosis of at least 50% had developed in 22 femoropopliteal arteries (Table). When restenosis and reocclusion combined were considered treatment failures, the cumulative patency rate at follow-up was 53% (SE, 7%) after 1year, 33% (SE, 7%)after 2 years, and 30% (SE, 8%)after 3 years (Figure). New stenoses of at least 50% occurred in 12 limbs (24%) during follow-up in segments other than those initially treated. Nine of these lesions developed in limbs in which there was a simultaneous restenosis, and three occurred in limbs with a patent recanalization site. When restenoses of at least 50%, reocclusions, and new stenoses of at least 50% were combined and were
Vroegindeweij et a1
335
Volume 6 Number 3
Summary of Lesions Detected by Means of Color-flowDuplex US and Arteriography
Lesion Restenosis of 250% Occlusion New stenosis of 250%
Color-flow Duplex US
Arteriography
21 11 9
22
considered to represent treatment failure, the cumulative patency a t follow-up was 53% (SE, 7%) after 1 year, 32% (SE, 7%) after 2 years, and 29% (SE, 8%) after 3 years. In one patient with a calcified femoral artery, duplex US demonstrated restenosis of less than 50%. and arteriography demonstrated stenosis of 65%. Three new stenoses were not depicted at duplex US because of severe calcification in the distal femoral artery a t the site of the Hunter canal.
DISCUSSION Several studies of the long-term results of recanalization of the femoropopliteal artery with either coldor hot-tip techniques have recently been reported (1-19). Despite the abundance of literature on this subject, it is difficult to find consistency in the outcomes because most study results rely on nonstandardized clinical data and incidental arteriography examinations a t follow-up (2,7-9,11,15-17). This approach to the analysis of long-term findings of stenosis is insufficient, as clinical examination is insensitive in the detection of restenosis and therefore is less suited to identify substantial differences between treatment modalities. The use of more sensitive tests such as duplex US would increase the likelihood of significant differences being found in the outcomes of treatment with recently developed and experimental techniques. Moreover, comparative clinical trials might require less time and smaller sample sizes and would be less expensive to accomplish. From a statistical point of view, a
11 12
more accurate follow-up method would help detect an end point of patency in more patients; this increases the power of the comparative analysis in a study population. For instance, in the present study the rate of clinical deterioration a t 3 years was 54% compared with the 67% failure rate determined with duplex US. Arteriography is currently used to help detect recurrence of stenosis or occlusion, clinical deterioration, and decrease in ABI. However, arteriography is an invasive and expensive method and cannot be performed serially at follow-up. And although arteriography provides precise anatomic data, it does not provide functional information about blood flow nor does it indicate flow abnormalities. Moreover, the actual time required for development of recurrent or new disease remains somewhat unknown, as arteriography cannot be performed repeatedly. Until recently, there have been no reports to our knowledge of the longterm clinical, arteriography, and duplex US results after recanalization with standardized criteria of failure. In the present study we used the SVSASCVS guidelines for classification of clinical results and the wellestablished criteria for duplex US of Jager et a1 (22,24). The 3-year clinical patency rate of 46% after recanalization found in our study was comparable to the results in other studies (8,9). Recanalization results assessed by means of arteriography and duplex US were lower than those determined clinically, with differences in 3-year patency rates of 13% and 16%, respectively. Thus clinical assessment is less sensitive in the detection of recurrent disease
than examination with arteriography and duplex US. These findings are in agreement with those of Miller et a1 (23), who reported that restenosis was depicted on color-flow images before it was found by means of ABI. Different threshold values of the PSV ratio chosen to represent recurrent disease correlate with different severity of disease. A PSV ratio of a t least 2.0 correlates well with the early development of restenosis, while a higher PSV ratio, of 2.5 for instance, may better correlate with the recurrence of clinical symptoms (31). The results of the present study, as well as those of other studies, suggest that the recurrence of symptoms combined with a decrease in resting ABI is not a sensitive indicator of restenosis (20,211. One factor may be that the development of collateral vessels abolishes the return of symptoms in cases of recurrent disease. It is our experience that for a complete clinical evaluation, the postexercise ABI should be included, as it seems to help identify more recurrent lesions than does resting ABI alone. In the present study the decrease in ABI, from 0.89 0.11 to 0.78 f 0.17, after clinical deterioration was rather modest compared with the significant decrease, from 0.84 f 0.16 to 0.51 f 0.16, in postexercise ABI. Compared with duplex US, however, the postexercise ABI is still less sensitive in the detection of recurrent disease (91% vs 81%). Nevertheless, postexercise ABI still appears a useful method in the identification of restenosis after intervention when color duplex US is not available. A sensitive, noninvasive method is needed to achieve a detailed study of the development of restenosis and reocclusion. This technique should enable determination of the location of a lesion and allow for differentiation between recurrent and new disease; the latter is usually caused by progression of atherosclerosis a t areas other than the treatment site. Color duplex US meets these requirements and is ideally suited to
+
Journal of Vascular and Interventional Radiology
336
-
May-June 1995
assessment of the durability of patency after the treatment of lesions in the femoropopliteal system. Findings in our study of 91% sensitivity and 100% specificity compared with arteriography confirm its reliability. The comparable accuracy of duplex US and arteriography in the diagnosis and grading of femoropopliteal lesions has been documented previously (20,23-27). In addition, duplex US can help differentiate between restenosis and reocclusion, and the severity of stenoses can be gauged and monitored with respect to their progression. We believe that regular follow-up with duplex US in patients who have undergone femoropopliteal recanalization is useful. Balloon angioplasty of stenotic lesions is simpler and has been associated with a higher procedural success rate than recanalization of occlusions (9). Moreover. the best long-term results with endovascular procedures in femoro~o~liteal disease have been obtained in patients with stenotic rather than occlusive disease (8,9, 16,19). Johnston et a1 (9) reported a success rate of 62% after 3 years in patients with stenoses, but that rate was only 46% in patients with occlusions. Also, Capek et a1 (8) and Gallino et a1 (19) reported long-term patency rates that were higher for stenoses than for complete occlusions (8,19). In our study, duplex US depicted a restenosis (21) considerably more often than a reocclusion (11)during follow-up. The importance of differentiating between restenosis and reocclusion was recently emphasized by Whyman et al (32), who said that the expected long-term outcome of reintervention was better for stenoses than for occlusions and that high-grade stenoses often progress into complete occlusions. Previous studies of the natural history of the development of femoropopliteal disease after endovascular procedures have demonstrated that restenoses occur much more frequently than progression of disease at the nontreated segments u
A
.
(33). New stenoses were detected in our study in nine patients by means of duplex US, and recurrent lesions were found in 32 patients. Differentiation of recurrent lesions in treated segments from new obstructions was achieved by arbitrarily dividing the superficial femoral artery into four segments and the popliteal artery into two segments. This scheme provided a way to record the location of the lesions demonstrated a t arteriography; a t follow-up with duplex US, the corresponding segments then could be identified with a ruler. This method of measurement in corresponding segments allowed us to compare disease identified by means of duplex US and arteriography and to use well-established criteria of severity. Because of the excellent agreement between duplex US and arteriographic findings, we now believe that reintervention in patients with high-grade, recurrent stenotic lesions may be planned without performance of arteriography for confirmation. This type of management of recurrent stenotic lesions will require less arteriography time, resulting in a more cost-effective use of health-care resources. We believe that the application of duplex US in the objective assessment of late results of percutaneous interventions should be expanded. This conclusion is in disagreement with the findings of Miller et a1 (231, who reported that none of their patients developed a late occlusion without an antecedent decrease in ABI; therefore, they stated their support for the use of ABI instead of the more expensive duplex US. Sensitivity with ABI was only 61% in our study, a finding which suggests that this method is not reliable in the detection of restenosis. However, we agree with Miller et a1 (23) that early detection of an increased PSV, which represents stenosis of at least 50%, a t duplex US without clinical complaints is not an indication for reintervention. However, during the follow-up period these stenoses could develop into high-grade stenoses or occlusions and could lead to clinical
deterioration. As stated previously, knowledge of impending occlusion with clinical complaints might allow for prevention in the preocclusive stage and might result in a better patency compared with occlusions (8,9,19). Our observations have strengthened our belief that use of a duplex US protocol after intervention is highly beneficial and may help improve overall patency. Especially during the first year of follow-up, we recommend a strict protocol, based on our observation that 23 of the 32 recurrences of stenosis and occlusion in this study were diagnosed in this period. After the first year, larger intervals between duplex US examinations are justified because of the lower incidence of recurrent lesions. In some cases in our study, there was no agreement between duplex US and arteriographic findings. Severe calcification of vessels often precludes visualization of stenotic lesions at duplex US because of the ultrasound beam attenuation. Calcification caused one restenosis and three new stenoses in Hunter canal to go undetected a t duplex US; these lesions were diagnosed only with arteriography. Despite these discrepancies, there was no statistically significant difference between the overall findings with arteriography and with duplex US. In conclusion, findings a t clinical examination, based on clinical symptoms and ABI, were insufficient in the detection of restenoses and reocclusions. Patency rates after 3 years of follow-up were comparable as determined with arteriography and duplex US. Duplex US has the advantage of being a noninvasive method that can be performed serially to help identify the development of restenosis. Timely identification by means of serial duplex US combined with reintervention may improve the overall results of recanalization of occlusions in the femoropopliteal arterial system. References 1. Berengoltz-Zlochin SN, Westerhof PW, Mali WPTM, et al. Nd:YAG la-
Vroegindeweij et al • 337 Volume 6 Number 3
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
ser-assisted angioplasty in femoropopliteal artery occlusions: "hot" versus "cold" recanalization with transparent contact probe. Radiology 1992; 182:409-414. Tobis JM, Conroy R, Stuart-Deutsch L, et al. Laser-assisted versus mechanical recanalization of the femoral arterial occlusions. Am J Cardiol 1991; 68:1079-1086. Luft CH, Horvath W, Oertl M, Haidinger D. Laser- versus rotationsangioplastie bei der rekanalisation chronischer femeropoplitealer arterien-verschliisse. Fortschr Rontgenstr 1993; 158:53-58. Odink HF, de Valois HC, Eikelboom BC. Femoropopliteal arterial occlusions: laser-assisted versus conventional percutaneous transluminal angioplasty. Radiology 1991; 181:61-66. Delcour CP, Wery D, Bank WO, Golzarian J, Struyven J. Arterial recanalization: hot-tip laser-assisted angioplasty versus hydrophilic guide-wire recanalization (abstr). Radiology 1991; 18l(P):266. Tobis J, Smolin M, Mallery J, et al. Laser-assisted thermal angioplasty in human peripheral artery occlusions: mechanism of recanalization. JAm Coll Cardiol1989; 13:15471554. Huppert PE, Duda SH, Helber U, Kansch KR, Clausen CD. Comparison of pulsed laser-assisted angioplasty in femoropopliteal artery occlusions. Radiology 1992; 184:363-367. Capek P, McLean GK, Berkowitz HD. Femoropopliteal angioplasty: factors influencing long-term success. Circulation 1991; 83(suppl 1):70-80. Johnston KW, Rae M, Hogg-Johnston SA, et al. Five-year results of a prospective study of percutaneous transluminal angioplasty. Ann Surg 1987; 206:403-413. Murray RR Jr, Hewes RC, White RI Jr, et al. Long-segment femoropopliteal stenoses: is angioplasty a boon or a bust? Radiology 1987; 162:473476. Hewes RC, White RI Jr, Murray RR, et al. Long-term results of superficial femoral artery angioplasty. AJR 1986; 146:1025-1029. Belli AM, Cumberland DC, Procter AE, Welsh L. Total peripheral ar-
13.
14.
15.
16.
17.
18. 19.
20.
21.
22.
23.
tery occlusions: conventional versus laser thermal recanalisation with a hybrid probe in percutaneous angioplasty-results of a randomized trial. Radiology 1991; 181:57-60. Zocholi G, Hotker U, Hake U, Storkel S, Schild H. Gefassrekanalisation mit einem thermo-kauter-katheter: in vitro-erfahrungen mit einem neuen kathetersysteem. Fortschr Rontgenstr 1988; 149:526528. Bonn J. Clinical utility oflaser recanalization in occluded peripheral arteries. Radiology 1991; 178:323325. Sanborn TA, Cumberland DC, Greenfield AJ, Welsh CL, Guben JR. Percutaneous laser thermal angioplasty: initial results and 1year follow-up in 129 femoropopliteal lesions. Radiology 1988; 168:121-125. Krepel VM, van Andel GJ, van Erp WFM, Breslau PJ. Percutaneous transluminal angioplasty of the femoropopliteal artery: initial and long-term results. Radiology 1985; 156:325-328. Martin EC, Fankuchen EI, Karlson KB, et al. Angioplasty for femoral artery occlusion: comparison with surgery. AJR 1981; 137:915-919. Gould DA. Thermal angioplasty using a high frequency ablator. J Intervent Radiol1990; 5:147-148. Gallino A, Mahler F, Probst P, Nachbur B. Percutaneous transluminal angioplasty of the arteries of the lower limbs: a 5 year follow-up. Circulation 1984; 70:619-623. Kinney EV, Bandyk DF, Mewissen MW, et al. Monitoring functional patency of percutaneous transluminal angioplasty. Arch Surg 1991; 126;743-747. Salles-Cunha SX, Andros G, Dulawa LB, Harris RW, Oblath RW. Changes in peripheral hemodynamics after percutaneous transluminal angioplasty. J Vasc Surg 1989; 10:338-342. Rutherford RB, Flanigan DP, Gupta SK, et al. Suggested standards for reports dealing with lower extremity ischemia. J Vasc Surg 1986; 4:80-94. Miller BV, Sharp WJ, Shamma AR, Kresowik TF, Petrone S, Corson JD. Surveillance for recurrent stenosis after endovascular procedures. Arch Surg 1991; 126:867872.
24. Jager KA, Ricketts HJ, Strandness DE Jr. Duplex scanning for the evaluation of lower limb arterial disease. In: Bernstein EF, ed. Noninvasive diagnostic techniques in vascular disease. Philadelphia, Pa: Mosby-Year Book, 1985; 619-631. 25. Sachs D, Robinson ML, Marinelli DL, Perlmutter GS. Evaluation of the peripheral arteries with duplex US after angioplasty. Radiology 1990; 176:39-44. 26. Kohler TR, Nance DR, Cramer MM, et al. Duplex scanning for diagnosis of the aortoiliac and femoropopliteal disease: a prospective study. Circulation 1987; 76:1074-1080. 27. Polak JF, Karmel MI, Mannick JA, et al. Determination of the extent of lower-extremity peripheral arterial disease with color-assisted duplex sonography: comparison with angiography. AJR 1990; 155:10851089. 28. Adar R, Critchfield GC, Eddy DM. A confidence profile analysis of the results of femoropoliteal percutaneous transluminal angioplasty in the treatment of lower extremity ischemia. J Vasc Surg 1989; 10:57-67. 29. Jeans WD, Armstrong S, Cole SEA, Horrocks M, Baird RN. Fate of patients undergoing transluminal angioplasty for lower limb ischemia. Radiology 1990; 177:559-564. 30. Landis JR, Koch GG. The measurements of observer agreement for categorical data. Biometrics 1977; 33:159-174. 31. Legemate DA, Teeuwen C, Hoeneveld H, Ackerstaff RGA, Eikelboom BC. Spectral analysis criteria in duplex scanning of aortoiliac and femoropopliteal arterial disease. Ultrasound Med Biol1991; 17:769-776. 32. Whyman MR, Ruckley CV, Fowkes FGR. A prospective study of the natural history of femoropopliteal artery stenosis using duplex ultrasound. Eur J Vasc Surg 1993; 7:444-447. 33. Adams DH, Schoen FJ. Contemporary concepts in atherosclerosis pathology. In: White RA, ed. Atherosclerosis and arteriosclerosis: human pathology and experimental animal methods and models. Boca Raton, Fla: CRC, 1989; 49-86.