This study summarizes the management of 24 patients who experienced coronary artery perforation as a complication of percutaneous coronary intervention (PCI) at a hospital in Iraq from 2009-2016. The majority of perforations involved the left anterior descending artery and were classified as Type II or III, requiring sealing with covered stents. Thirteen patients also required drainage of pericardial effusions. All perforations were immediately diagnosed and treated, with no patients requiring surgery or experiencing mortality. The low rate of coronary artery perforation complications in this study, primarily managed using covered stents, demonstrates the effectiveness of the approaches used at this hospital.

1. Original Article
Coronary artery perforation complicating
percutaneous coronary intervention
Aram J Mirza1
, Abdulsalam Y Taha2
, Jaafar S Aldoori1
,
Jawad M Hawas1
and Kawa W Hassan1
Abstract
Background: Coronary artery perforation is a rare but serious complication of percutaneous coronary interventions.
We aimed to evaluate the management of coronary artery perforation in Sulaimaniyah, Iraq.
Methods: A retrospective review of our medical records from 2009 to 2016 identified 24 patients (15 males, 9 females)
with coronary artery perforation. Mean age was 60 Æ 9.2 years (range 40–74 years). Standard diagnostic angiography or
percutaneous interventions were performed. Coronary artery perforation was diagnosed by worsening of symptoms,
hypotension, or angiographic evidence of type I (extraluminal crater), II (myocardial or pericardial blushing), or III
(contrast streaming or cavity spilling) perforation. Stenosis was graded as >85%, 60%–85%, or < 60%. Once coronary
artery perforation was diagnosed, heparin was reversed, antiplatelets were stopped, and pericardial effusions were
aspirated. Type II and III coronary artery perforations were sealed using covered stents or repeated brief balloon
inflations.
Results: The most frequently injured artery was the left anterior descending (n ¼ 14, 58.3%). Type II and III coronary
artery perforations constituted the majority (n ¼ 18, 75%). Thirteen (54.2%) patients had severe coronary stenosis.
Perforations were caused by stents (n ¼ 10), angioplasty wires (n ¼ 8), and balloons (n ¼ 6). Fifteen perforations were
sealed with covered stents, 2 by balloon inflations, and 7 resolved spontaneously. Pericardial effusion was drained in 13
(54.2%) patients. No patient required surgery, and none died.
Conclusion: The low rate and early management of coronary artery perforations, mainly by covered stents, were the
hallmarks of this study.
Keywords
Coronary angiography, Coronary artery disease, Coronary vessels, Percutaneous coronary intervention, Stents, Vascular
system injuries
Introduction
The gold-standard therapy for coronary artery disease
(CAD) nowadays is percutaneous coronary interven-
tion (PCI).1
Millions of patients undergo diagnostic
coronary angiography and PCI by expert cardiologists
in cardiac centers all over the world, with a high degree
of safety. Nevertheless, complications do occur. Most
are vascular access site complications. Acute myocar-
dial infarction, coronary artery perforation (CAP),
stroke, and death are serious but fortunately rare com-
plications, with an incidence of approximately 1%.2
CAP is more common with PCI than with diagnostic
coronary angiography.2
Slemani Cardiac Hospital is
the first and only public cardiac center in
Sulaimaniyah, Iraq. The departments of cardiology
and cardiac surgery serve the Kurdistan region as well
as other Iraqi governorates. The aim of this study was
Asian Cardiovascular & Thoracic Annals
2018, Vol. 26(2) 101–106
ß The Author(s) 2018
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0218492318755182
journals.sagepub.com/home/aan
1
Department of Cardiology, Slemani Cardiac Hospital, Sulaimaniyah,
Region of Kurdistan, Iraq
2
Department of Thoracic and Cardiovascular Surgery, School of
Medicine, Faculty of Medical Sciences, University of Sulaimaniyah,
Sulaimaniyah, Region of Kurdistan, Iraq
Corresponding author:
Abdulsalam Y Taha, University of Sulaimaniyah, Mamostayan Street 112,
Road 33, House 5, Sulaimaniyah 46001, Region of Kurdistan, Iraq.
Email: salamyt_1963@hotmail.com

2. to review the management of this rare yet serious com-
plication of PCI in our hospital.
Patients and methods
This was a retrospective study of 24 patients (15 men
and 9 women) with CAP that developed during PCI for
CAD over an 8-year period (2009–2016). The medical
records of all patients who had CAP in the study period
were reviewed for age, sex, history of chest pain preci-
pitated by exertion and relieved by rest, shortness of
breath, orthopnea, paroxysmal nocturnal dyspnea and
leg edema, family history and risk factors of CAD,
such as diabetes mellitus, hypertension, and smoking.
Every patient was thoroughly investigated by plain
chest radiography, electrocardiography, and
echocardiography.
All patients received aspirin 100 mgÁday 10 days
prior to PCI (unless pretreated) and were loaded with
clopidogrel 300 mg or prasugrel 60 mg the day before
PCI. A bolus dose of 5000–10,000 units of unfractio-
nated heparin was given during the procedure.
Standard coronary angiography Æ PCI was performed
under local anesthesia via the transfemoral or transra-
dial approach (according to operator preference). The
severity of coronary artery stenosis was graded as A
(>85%), B (60%–85%), or C (<60%).
CAP was detected by worsening of symptoms, per-
sistent hypotension, and/or angiographic findings. It
was categorized as type I (extraluminal crater), II (myo-
cardial or pericardial blushing), or III (contrast stream-
ing or cavity spilling). The causes of CAP were
angioplasty wires, balloons, or stents. The treatment
was based on type of CAP, hemodynamic state of the
patient, and the severity of pericardial effusion (PE).
Stable patients with mild PE (<100 mL) and type I
CAP required no pericardial aspiration, just reversal
of heparin by protamine, cessation of antiplatelets,
and close clinical and echocardiographic monitoring.
Unstable patients with moderate (100–400 mL) to
severe (>400 mL) PE and type II/III CAP required fur-
ther measures such as urgent subxiphoid pericardiocen-
tesis, a few days of catheter (6F pigtail) drainage, and
sealing of the perforation by either balloon inflation or
covered stent (Abbott Vascular) placement. The bal-
loon was inflated briefly (up to 4 min) to avoid myocar-
dial ischemia but repeatedly until bleeding ceased.
Coronary artery bypass grafting (CABG) was reserved
for patients who failed to respond to the previous meas-
ures and therefore, in every case of CAP, the cardiac
surgeon was informed. Emergency surgery in such
patients carries a high risk, but fortunately, it is rarely
needed. In this study, a conservative approach was used
to describe all treatment measures short of balloon
inflation and covered stent placement.
The patients could be discharged home on dual anti-
platelet therapy usually within a few days of sealing the
perforation. Clinical and echocardiographic follow-up
was arranged after one month to deal with any
complication.
Results
During the 8-year study period, 15,430 patients had
diagnostic coronary angiography Æ PCI. Hence, the
rate of CAP in our hospital was 0.16%. Table 1 shows
the age and sex distribution of the patients. The male/
female ratio was 15:9 (1.7:1). Ages ranged from 40 to 74
years with a mean of 60 Æ 9.2 years; the peak was in the
7th decade. Table 2 shows the risk factors according to
sex. All 10 smokers were male. Table 3 lists the severity
of coronary artery stenosis (A, B, C) and type of CAP (I,
II, III) according to sex. More than half of the patients
(54.2%) had severe coronary artery stenosis grade A
(>85%) and the majority (75%) had type II or III
CAP. Table 4 shows the severity of CAD and type of
Table 1. Age and sex distribution in patients with coronary
artery perforation after percutaneous coronary interventions.
Age (years) Males Females Total
40–50 3 2 5
51–60 3 5 8
61–70 8 1 9
71–80 1 1 2
Total 15 (62.5%) 9 (37.5%) 24 (100%)
Table 3. Type of coronary artery perforation and severity of
stenosis, according to the sex of the patient.
Sex
Severity of stenosis Type of CAP
TotalA B C I II III
Male 8 6 1 4 3 8 15
Female 5 3 1 2 2 5 9
Total 13 9 2 6 5 13 24
Table 2. Major risk factors for coronary artery perforation,
according to the sex of the patient.
Sex DM HT DM þ HT Smoking DM þ HT þ Smoking
Males 4 4 3 10 1
Female 2 2 2 0 0
Total 6 6 5 10 1
DM: diabetes mellitus; HT: hypertension.
102 Asian Cardiovascular & Thoracic Annals 26(2)

3. CAP. Patients with severe CAD (grade A) had mostly
type III CAP (10/13) while those with mild lesions
(grade C; 2/2) developed just type I CAP. Table 5
shows the diseased arteries, cause of perforation (wire,
balloon, stent), and treatment options (conservative,
balloon, covered stent). The most frequently injured
artery was the left anterior descending artery (58.3%).
A balloon was the least frequent cause while stents and
angioplasty wires accounted for most CAP (10 and 8,
respectively). Seven (29.2%) patients were managed
conservatively (Figure 1); 5 of them had right coronary
artery perforations caused by angioplasty wires on 4
occasions. The majority of perforations (62.5%) were
repaired with covered stents (Figure 2) and 2 were man-
aged by balloon inflation. All CAP were immediately
diagnosed and repaired by the cardiology team.
Table 6 gives the severity and treatment of PE; 11
patients were just observed and 13 (54.2%) were mana-
ged by aspiration and catheter drainage for a few days.
The volume of PE ranged from 100 mL to 550 mL; for-
tunately, no patient developed cardiac tamponade and
none needed emergency CABG. The duration of hospi-
talization was 24–96 h. Two patients were lost to follow-
up and the others were followed up for 2 years (Table 7).
All did well for 2 years, except one who developed
thrombosis of the covered stent once. No patient died.
Discussion
The rate of CAP in our study was 0.16% which is sub-
stantially lower than the worldwide published rate of
0.43%.3
This could be due to the availability of more
advanced medical technology in centers abroad, which
enables interventional cardiologists to treat more com-
plex lesions,4
leading to more CAP. To the best of our
knowledge, this is the largest study that addresses CAP
as a complication of diagnostic coronary angiography
or PCI in Iraq. Haji5
studied 80 patients with chronic
total occlusion who underwent PCI in 2 Iraqi cardiac
centers over one year, and reported 2 (2.5%) cases of
CAP; unfortunately, the details were not given. Apart
from this, we could not find any other local publica-
tions on CAP. The rarity of this complication is the
most likely explanation. On the other hand, the more
common but less serious vascular access complications
of diagnostic coronary angiography or PCI were well
studied by Majeed and colleagues6
and also by Al-
Marayati and Al-Mosawi.7
It is thought that the more risk factors the patient
has the more he is liable to develop complications of
Figure 1. Coronary angiography in left anterior oblique view,
showing a coronary artery perforation (Ellis type II) in the right
coronary artery with mild pericardial effusion. The patient was
managed conservatively.
Table 5. Diseased arteries, causes and treatment options for coronary artery perforation.
Artery
Cause of coronary artery perforation Treatment
Wire Balloon Stent Total Conservative Balloon Covered stent Total
Right coronary 4 1 1 6 5 1 0 6
Left anterior descending 2 4 8 14 1 1 13 15
Left circumflex 2 1 1 4 1 0 2 3
Total 8 6 10 24 7 2 15 24
Table 4. Severity of stenosis and type of coronary artery
perforation.
Severity of stenosis
Type of coronary artery perforation
I II III Total
A (>85%) 0 3 10 13
B (60%–85%) 4 2 3 9
C (<60%) 2 0 0 2
Total 6 5 13 24
Mirza et al. 103

4. PCI, probably due to worsening of the underlying dis-
ease. Al-Najjar8
found that the rate of complications
was nil in patients with no risk factor but reached
33.3% in patients with 3–5 risk factors. In our study,
20 (83.3%) patients had 1–3 risk factors. Ten (41.7%)
male patients were smokers, similar to the findings of
Lee and colleagues9
(44.4%). Both age and female sex
are considered risk factors for CAP.2,3,10
The mean age
of our patients was 60 years and the peak was in the 7th
decade. The risk of PCI complications is higher in
females due to more comorbidities, more risk factors,
smaller body size, and smaller blood vessels.7
Unlike
other studies,2,3,7,10
most of our patients were male,
with a male/female ratio of 1.7:1. Al-Najjar8
noted
that hypertension was associated with the highest rate
of PCI complications (27.8%), possibly due to the link
between hypertension and arterial dissection, the pre-
cursor of most complications including perforation. In
our study, a quarter (6/24) of the patients were hyper-
tensive. Al-Daghir and colleagues11
found that perfor-
ation occurred in 13.2% of diabetic patients compared
to 3.4% in the non-diabetic group. Similarly, in the
study by Lee and colleagues,9
40.4% of patients were
diabetic. In our study, 6 (25%) patients were diabetic
and 5 (20.8%) were both hypertensive and diabetic;
thus 11 (45.8%) patients had diabetes mellitus. One
patient was hypertensive, diabetic, and a smoker at
the same time.
Figure 2. (a) Coronary angiography in left anterior oblique 40-10 angulation view, showing a coronary artery perforation (Ellis type
III) in the left anterior descending artery with a moderate pericardial effusion. (b) The perforation was sealed by a covered stent and
400 mL of pericardial fluid was aspirated. DES: drug-eluting stent.
Table 7. Follow-up details.
Treatment
No. of
Patients
Lost to
follow-up
Patients followed
up for 2 years
No
complication Complication
Conservative 7 0 7 7 0
Balloon inflation 2 1 1 1 0
Covered stent 15 1 14 13 1 (in-stent thrombosis)*
Total 24 2 22 (100%) 21 (95.5) 1 (4.5%)
*Detected 6 months post-procedure; clopidogrel has been discontinued by the patient when he developed gastric erosion.
Table 6. Severity and treatment options of pericardial effusion, according to the sex of the patient.
Sex
Pericardial effusion Treatment
Mild (<100 mL) Moderate (100–400 mL) Severe (>400 mL) Aspiration Æ drainage Observation Total
Male 7 2 6 8 7 15
Female 4 2 3 5 4 9
Total 11 4 9 13 11 24
104 Asian Cardiovascular & Thoracic Annals 26(2)

5. CAP is defined as an anatomical breach in the wall
of a coronary vessel due to penetration of the 3 layers
of the vessel wall, resulting in extravasation of blood or
dye into the pericardium, myocardium, or adjacent car-
diac chamber or vein.4
Extravasation of blood into the
left or right ventricle, if not massive, has minimal imme-
diate clinical consequences. Extravasation into the
myocardial space can yield a myocardial hematoma
(fortunately very rare),12
with potentially serious con-
sequences, whereas bleeding into the pericardial space
can have catastrophic consequences.4
CAP can be clas-
sified according to location (main vessel, distal artery,
collateral vessel) and severity (using the Ellis classifica-
tion of 3 types).13
The etiology of CAP following PCI is
multifactorial. Beside the risk factors of age, female sex,
diabetes, hypertension, and smoking, anatomical com-
plexity, the equipment used, and the skill of the oper-
ator are also factors. Rates of CAP are potentially
higher when complex calcified lesions are treated.1,3,10
Diagnostic coronary angiography is rarely complicated
by CAP.2
Atherectomy devices, oversized compliant
balloons, and stiff and hydrophilic wires are among
the devices known to produce more CAP.1,3,8,9,14
Guidewire-related perforations are most frequently
encountered as result of accidental migration of the
wire deep into the coronary microvasculature or use
of aggressive wires that penetrate the adventitia.4
Deep cannulation of a coronary artery, forceful con-
trast injection, and over-sizing of balloons and stents
must be avoided.2,3
More than half of our patients
(54.2%) had severe coronary artery stenosis (>85%)
which could be one of the causative factors. Stents
were responsible in 10 cases (41.7%), similar to the
finding of Lee and colleagues.9
Wires were responsible
for 8 CAP in our study, and balloons in 6. Lee and
colleagues9
also noted that the left anterior descending
artery was the most frequently injured vessel, as in our
study. Likewise, type III CAP predominated in our
study. Morbidity and mortality are directly related to
the Ellis classification of CAP, being higher in type III.
Type I perforations rarely lead to tamponade compared
to type III.15
It is crucial to immediately diagnose CAP during
PCI, clinically or angiographically, and to promptly
initiate therapy. The optimum therapy for CAP is still
controversial, but it is vital to seal the perforation and
prevent further consequences.4
In this series, all CAP
were immediately diagnosed and repaired by the cardi-
ology team. Although PE developed in all patients, it
was moderate to severe, requiring aspiration and drain-
age only in 13 (54.2%) patients; none progressed to
cardiac tamponade. Immediate reversal of anticoagula-
tion is a key element in any perforation that might lead
to cardiac tamponade or a large myocardial hema-
toma.4
Prolonged balloon inflation may result in
intracoronary thrombosis, myocardial infarction, and
death.1
To relieve chest pain and avoid ischemia of
the distal area during balloon inflation, a micro-cathe-
ter over another guidewire is positioned distal to site of
perforation, and the patient’s own arterial blood is
injected (micro-catheter distal perfusion technique).14,15
Two of our patients were successfully managed by brief
(up to 4 min) repeated balloon inflation to avoid myo-
cardial ischemia. The use of covered stents has revolu-
tionized the management of coronary perforation.1
The
polytetrafluoroethylene-covered stent is composed of a
membrane sandwiched between 2 metallic stents.9
These stents prevent blood leakage between the stent
struts, and are highly successful.13
However, they have
some drawbacks such as lack of elasticity, difficult
deployment in calcified vessels, and a higher rates of
stent re-stenosis (reaching 31.6% at 6-month follow-
up) and in-stent thrombosis (15.6%).9
In the present
study, covered stents were the technique of choice to
seal CAP, with only one instance of in-stent thrombosis
(4.5%), much lower than the reported rate. Distal or
end-artery perforations can be sealed using thrombo-
genic metallic coils, polyvinyl alcohol, collagen foam,
intra-arterial thrombin, or autologous fat tissue aspi-
rated from the patient’s groin. These techniques have
been shown to be safe, effective, and inexpensive, espe-
cially for small vessel perforations, although these may
be less predictable and more technically challenging.4
Although CABG is an excellent treatment for type III
perforation, the time needed to prepare the operating
room might be too long.1
CABG can be undertaken
when a coated stent and pericardial drainage have not
corrected the initial hemodynamic compromise.1
Emergency CABG is especially important for patients
with CAP and prior CABG surgery. Such patients tend
to develop loculated hematoma compressing a cardiac
chamber rather than global pericardial tamponade;
hence surgical evacuation may be necessary.13
When
we started our work, we discussed the necessity of sur-
gical backup before we decided to perform PCI because
it definitely carries a risk of perforation. This study
clearly shows that CAP can be correctly diagnosed
and properly treated by the operating cardiologist in
most cases. However, we still believe that a cardiac sur-
geon should be available to give a hand if needed. Some
cases of CAP type III with bleeding into the pericar-
dium and potential cardiac tamponade may fail to
respond to nonoperative measures such as pericardio-
centesis, balloon inflation, and covered stent placement;
thus emergency CABG would be the last resort.
Causes of death after CAP include acute cardiac
tamponade, acute myocardial infarction, and myocar-
dial hematoma.1,4
The mortality is directly related to
the Ellis class and ranges between 0.3% in type I to
21.2% in type III.3
Al-Najjar8
reported that one patient
Mirza et al. 105

6. out of 115 (0.9%) died due to coronary thrombosis 72 h
after coronary artery rupture during PCI that was suc-
cessfully stented, in a 4-year study from Saudi Arabia.
In our study, there was no mortality and CAP were
immediately diagnosed and repaired by the cardiology
team without CABG. Nevertheless, surgical back up
should not be forgotten. We agree with the conclusion
of Dash:15
‘‘No matter how experienced the operator is,
complications can never be completely avoided, rather
they would teach humility and be a continuous source
of education’’; and also with Klaas and colleagues:16
‘‘Whether an injury is iatrogenic does not matter;
what matters is whether an injury was caused by
negligence’’.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
References
1. Piraino D, Dendramis G, Buccheri D, et al. Coronary
artery perforation: how to treat it? Cor et Vasa 2015; 57:
e334–40. Available at: http://www.sciencedirect.com/sci-
ence/article/pii/S0010865015000624. Accessed January 03,
2018.
2. Gu¨ nes¸ H, Sahin O, Bolayir HA, Tatlisu MA, Kivrak T and
Karaca I. Rare complication of diagnostic coronary angi-
ography: perforation. Int J Cardiovasc Acad 2017; 3: 45–7.
Available at: http://dx.doi.org/10.1016/j.ijcac.2017.01.003.
Accessed January 03, 2018.
3. Jurado-Roman A, Garcia-Tejada J, Hernandez-
Hernandez F, et al. Coronary artery perforation: don’t
rush, IVUS may be useful. Rev Port Cardiol 2015; 34:
623e1–3.
4. DePersis M, Khan SU, Kaluski E and Lombardi W.
Coronary artery perforation complicated by recurrent car-
diac tamponade: a case illustration and review. Cardiovasc
Revasc Med 2017; 18: S30–S34.
5. Haji GF. Success rate of percutaneous coronary interven-
tion of chronic total occlusion in Ibn Al-Bitar Hospital for
Cardiac Surgery and Al-Nassyeria Cardiac Center.
Al-Kindy Col Med J 2012; 1: 75–82.
6. Majeed SM, Al Saffar HB and Al-Marayati AN.
Complication Following percutaneous coronary interven-
tion via the femoral artery experience in lraqi Center for
the Heart Disease and lbn Al- Bitar Hospital for cardiac
surgery. Fac Med Baghdad 2016; 4: 325–9. Available
at: https://www.iasj.net/iasj?func¼fulltext&aId¼118035.
Accessed January 03,2018.
7. Al-Marayati AN and Al-Mosawi N. In-hospital outcome
and complications of percutaneous coronary intervention
in acute coronary syndrome, Gender Differences. Fac
Med Baghdad 2008; 50: 420–4243. Available at: https://
www.iasj.net/iasj?func¼fulltext&aId¼958. Accessed
January 03,2018.
8. Al-Najjar H. Impact of risk factors on complications of
coronary intervention in non-ST-elevation acute coron-
ary syndrome. Iraqi Postgrad Med J 2009; 8: 105–10.
Available at: https://www.iasj.net/iasj?func¼fulltext&
aId¼47961. Accessed January 03,2018.
9. Lee WC, Hsueh SK, Fang CY, Wu CJ, Hang CL and
Fang HY. Clinical outcomes following covered stent for
the treatment of coronary artery perforation. J Interven
Cardiol 2016; 29: 569–575.
10. Lee MS, Shamouelian A and Dahodwala MQ. Coronary
artery perforation following percutaneous coronary inter-
vention. J Invasive Cardiol 2016; 28: 122–131.
11. Al-Daghir HA. Percutaneous revascularization of
chronic total occlusion of diabetic patients at Iraqi
center for heart diseases, a single center experience
2012. Muthanna Med J 2015; 2: 76–82.
12. Kufner S, Cassese S, Ndrepepa G, Kastrati A and Fusaro
M. Diagnosis and management of intra-myocardial
hematoma after coronary artery perforation: case studies
of the Deutsches Herzzentrum Mu¨ nchen. Coron Artery
Dis 2016; 27: 327–330.
13. Karatasakis A, Akhtar YN and Brilakis ES. Distal cor-
onary perforation in patients with prior coronary artery
bypass graft surgery: the importance of early treatment.
Cardiovasc Revasc Med 2016; 17: 412–447.
14. Yamamoto S, Sakakura K, Funayama H, Wada H,
Fujita H and Momomura S. Percutaneous coronary
artery bypass for type 3 coronary perforation. JACC
Cardiovasc Interv 2015; 8: 1396–1398.
15. Dash D. Complications encountered in coronary chronic
total occlusion intervention: prevention and bailout.
Indian Heart J 2016; 68: 737–746.
16. Klaas PB, Berge KH, Klaas KM, Klaas JP and Larson
AN. When patients are harmed, but are not wronged:
ethics, law, and history. Mayo Clin Proc 2014; 89:
1279–1286.
106 Asian Cardiovascular & Thoracic Annals 26(2)