CARDIOVASCULAR JOURNAL OF AFRICA • Vol 21, No 6, November/December 2010
314
AFRICA
Long coronary lesions comprise up to 20% of current inter-
ventional practice and are considered difficult both technically
and in terms of achieving successful clinical outcomes.
1,14
To
date, both single long stents and multiple contiguous stents
have been used for the successful treatment of coronary artery
lesions.
5,15-17
Stent length was previously an independent predictor of
restenosis; however, the use of DES has greatly attenuated
this relationship. As a result, long DES tend to be selected for
complete lesion coverage, but if this approach is not possible
or a residual segment of the lesion is left uncovered, additional
stenting is considered, with some overlap to eliminate the risk of
a residual stent gap.
18,19
Some reports have shown an increased
rate of peri-procedural myonecrosis in overlapping stents, which
may be a limitation of this approach.
9,20
A pooled analysis of five
clinical trials showed that BMS overlap was associated with an
increased incidence of MI and total MACE that was not apparent
for DES overlap.
11
On the other hand, Finn
et al
. reported that, in a histological
analysis in an animal study, overlapping DES further delayed the
arterial healing and promoted inflammation compared to over-
lapping BMS.
21
Therefore they concluded that patients receiving
overlapping DES needed more frequent follow up than patients
with non-overlapping stents. It is noteworthy that many studies
show comparable results after PCI with homogeneous and heter-
ogeneous drug-eluting stents,
22,23
and a low rate of repeat revas-
cularisation irrespective of stent type, with no safety concerns at
medium-term follow up.
24,25
In 2006, Burzotta
et al
. distributed their report on a consecu-
tive series of 40 patients in Italy, treated with overlapping stents.
These stents were used to cover minor dissections and plaque
shifts, treat other contiguous lesions in the same vessel, or obtain
full lesion coverage. In a subgroup of patients, an appropriate
(length and size) similar type of DES was not available, so the
additional stent necessary to complete the procedure was a BMS
or another type of DES. In their study, 24 patients were treated
with overlapping homogeneous DES, eight with overlapping
heterogeneous DES, and eight with overlapping DES-BMS. In
their experience, three out of the 24 patients (12.5%) in the over-
lapping homogeneous DES group developed MACE: two, target
lesion revascularisations and one, coronary bypass surgery. No
MACE occurred in the overlapping heterogeneous DES group.
In the overlapping DES-BMS group, the rate of MACE was
50% (three target-vessel revascularisations and one death after
cardiogenic shock in a patient who developed ST-elevation MI
110 days after PCI).
22
As stated in the results section, we used homogeneous drug-
eluting stents in all 22 patients treated with overlapping DES,
which matches the Burzotta study. However, we used a combina-
tion of DES and BMS in a much higher number of patients than
in their registry (30 vs eight), and no MACE was experienced. In
our study, MACE only included MI after 13.2 months in a patient
treated with two overlapping Cyphers, yielding a MACE rate of
4.5%, compared to 0% in patients treated with DES-BMS. We
must point out that we used the definition of a CKMB rise two-
fold above baseline, although the new definition of MI is based
on troponin level.
26
Angiographic restenosis was detected in two of the six
patients who had undergone follow-up angiography. Since
we only performed follow-up angiography when considered
clinically appropriate (in 11.5% of cases), the true angiographic
restenosis rate in our cohort could not be established with
certainty, compared to Burzotta
et al
., who performed follow-
up angiography in more than 80% of cases. In their experience,
MACE was detected only in those patients who had undergone
follow-up angiography and patients without follow-up angiogra-
phy did not show MACE at nine months.
In the Burzotta study, total stent length was shorter in patients
treated with DES-BMS than DES-DES [39
±
16 mm in the
DES-BMS vs 50
±
10 mm in the overlapping Cypher and 42
±
7 mm the overlapping Taxus group (
p
=
0.09)].
22
Their results
also showed more intimal hyperplasia at the site of stent overlap
in DES-BMS overlaps than when homogeneous or heterogene-
ous DES were used. The higher late lumen loss translated into
a higher in-segment binary restenosis rate in lesion segments
covered with DES-BMS, therefore decreasing the possible
benefits associated with DES implantation.
In the setting of long dissections, multiple short stent place-
ments have proven to be equivalent to the use of long stents.
27
In
our practice, in four patients, all dissections were successfully
treated with overlapping DES-BMS (Table 4). In addition, as
stated in the results section, the additional bare-metal stents used
for coverage of residual lesions, both in the case of dissections or
for long lesions, were smaller and had larger diameters. Since the
number of endpoints was small in our study, we may not reach a
definite conclusion, but we hypothesise that the smaller lengths
and larger diameters of the bare-metal stents used for coverage of
residual lesions may have reduced the risk of MACE in the group
treated with overlapping DES-BMS.
On the other hand, we found that lesions were estimated to be
significantly shorter in the DES-BMS group, suggesting another
potential factor for the reduced risk of MACE. However, total
stent lengths were not significantly different between the two
groups. This may be explained by the fact that all the patients
treated with more than two stents were in the DES-BMS group,
and additional bare-metal stents were used for the coverage of
residual lesions.
Our results showed that bare-metal stents with relatively short
lengths and large diameters can be overlapped in the proximal
portion of a long drug-eluting stent for the coverage of residual
lesions when the DES length would not suffice for the coverage
of the total lesion, or in the case of proximal edge dissection.
In our practice, this procedure was both feasible and safe, with
no increased rate of late stent thrombosis, as opposed to the use
of two or more drug-eluting stents, which have been frequently
known to be associated with stent thrombosis, especially at the
site of overlapping stent struts.
Limitations of this study include the fact that it was a retro-
spective analysis of consecutive and non-randomised patients,
different stent types were used, and stent deployment techniques
varied between the operators. Also, these data reflect the current
clinical practice at our institution. Routine follow-up angiogra-
phy was therefore not performed, which blurred the true rate of
acquired restenosis within the study population.
Conclusion
In clinical practice, there are patients in whom the interventional
cardiologist is required to overlap two or more stents in order to
cover residual lesions or dissecting flaps of a long atherosclerotic
