Left Atrial Thromboendarterectomy in
Rheumatic Mitral Valve Patients
Noell Fasol1, Berit BODE1, Xue LI2, Anneli Seppälä Lindroos2 and Roland FASOL1*
1Department of Cardiac Surgery, China
2Department of Cardiology, China
Submission: November 14, 2018;Published: January 18, 2019
*Corresponding author: Roland FASOL, MD, Department of Cardiac Surgery, Tree Top Hospital, China
How to cite this article: Noell F, Berit B, Xue L, Anneli S L, Roland F. Left Atrial Thromboendarterectomy in Rheumatic Mitral Valve Patients. Open Access J Surg. 2019; 10(2): 555783. DOI: 10.19080/OAJS.2019.10.555783.
Background: Rheumatic mitral valvular heart disease is associated with the increased risk of significant left atrial enlargement and massive thrombus formation, although there are limited data on surgery and the prognosis of those patients.
Methods: We enrolled 16 patients (56.8±6 yrs) in rural China with massive thrombus formation in the enlarged left atrium (ø 61.5±6.5 mm) out of 120 consecutive patients with rheumatic mitral disease operated during a 6-month study period. Fourteen (87.5%) had concomitant diseases: tricuspid incompetence in 12, aortic stenosis in 3, and coronary artery disease in 3. The surgical technique of left atrial thromboendarterectomy included a careful blunt subendocardial dissection in order to separate ‘in toto’ the thick organized thrombotic material from the enlarged left atrial wall. Surgical volume reduction of the giant left atrium was performed using a circumferential atrioplasty of the posterior and inferior atrial wall. Early follow up was done after 3 months in the hospital, and after 6 and 12 months by telephone interview.
Results: There was no perioperative and early death in the 16 patients with a left atrial thromboendarterectomy and no neurological event. At one year follow up there was one confirmed postoperative death, 4 were lost to follow-up, but the remaining 11 patients describing their NYHA functional class as I and II and 3 patients even reported to be back at work.
Conclusion: Aggressive approach to left atrial thromboendarterectomy and atrioplasty for volume reduction in patients with rheumatic mitral stenosis and giant left atrium is safe, efficient and successful.
Keywords: Mitral Valve; Rheumatic Heart Disease; Left Atrium; Atrial Fibrillation
Rheumatic heart disease (RHD) seems to have been eradicated in western countries. However, it is not in the rest of the world . China has been struck by disasters, civil war and cultural revolution for the last centuries. Universities were closed, medical education stopped, and medical infrastructure was lost. As a result, we now see a high percentage of patients with rheumatic heart valve disease in newly opened heart centres. The Jilin Heart Hospital is one of the first private and non-profit heart centers in rural China. We initiated a novel social health care reimbursement model program for the Jilin Province-the JIXIN Program (Ji - Jilin/吉, Xin - Heart/心)-that allows a full cost coverage for cardiac diagnosis and treatment for the rural population of the Jilin province in our hospital. This allows for the first time a real medical treatment for the rural population. Since there is hardly any modern medical infrastructure available for patients in the countryside, most of our patients have ever been exposed to any kind of substantial medical diagnosis and treatment.
There are significant socio-economic disparities in China and a massive gap between urban and rural population groups . While the wealthier share of the Chinese population has benefited
from advanced health technologies and spending on health care, the poor have lost access to even the most essential services. In terms of rural-urban disparity across provinces, life expectancy drops parallel to a rising share of rural population . We know our patient population represents a rare group of patients, if compared to western countries were patients are normally exposed to state-of-the-art pre-operative diagnosis and treatment as well as post-operative follow-up and care.
Rheumatic heart disease is the most common cause of pathology in our patients coming for mitral valve surgery to our hospital. Atrial fibrillation enlarged left atria, severe pathological pulmonary hypertension and involvement of more than one valve is seen in most of our patients. Giant left atrium and massive thrombus formation is only seen in few. There are only several case reports [4,5] but no larger series found in the literature. This study describes a 6-month experience of 16 consecutive mitral valve patients who had left atrial thromboendarterectomy (LA-TEA) due to massive thrombus formation in the enlarged (giant) left atrium combined with a left atrioplasty for volume reduction.
This investigation was exempt from a formal Institutional
review board as it represents a retrospective analysis of our
patient data that are collected for quality analysis and purposes
other than research.
Patient records for all mitral valve operations were evaluated
from the department database for the 6-month study period.
Study inclusion criteria included all patients undergoing primary,
isolated or combined mitral valve surgery.
Preoperative, operative and early postoperative results
were assessed in 120 consecutive RHD patients who underwent
intervention on the mitral valve. Of these patients, 42 (35%)
underwent mitral repair, 78 (65%) mitral replacement and 90
(75%) had concomitant surgical procedures. Only 32 (26,6%)
patients had isolated mitral surgery. Rheumatic lesions were the
cause for surgery in all 120 patients (25%) out of 479 patients
operated during this period. The presenting symptom was
dyspnea at exertion and at rest. 106 patients (88.3%) were in
New York Heart Association (NYHA) functional class III and IV.
Preoperative and operative patient characteristics are detailed
in Table 1. In 16 (13.3%) of all RHD mitral patients massive
organized thrombi covering most of the left atrial cavity were
found (Figure 1A-C). Nine (56.3%) of these 16 patients had a
predominant mitral valve stenosis and 7 (43.7%) a mixed lesion.
Only 2 of the 16 patients had an isolated rheumatic mitral lesion,
14 (87.5%) had concomitant diseases: tricuspid incompetence in
12, aortic stenosis in 3, and coronary artery disease in 3. Some
patients had more than one concomitant procedure.
NYHA-New York Heart Association; ACC-aortic cross clamp (time/min); TVR-tricuspid valve repair; AVR-aortic valve replacement; BMI-body mass
In these 16 mitral patients (characteristics are detailed in
Table 2) a thromboendarterectomy and surgical atrioplasty for
volume reduction of the giant enlarged left atrium was performed
(Figures 2 & 3). In 6 (37.5%) of all 16 patients with left atrial
thrombi some history of a possible stroke or neurological event
was mentioned in their patient history. However, no detailed
documentation of the event was available, and no symptoms of a
neurological deficit could be observed.
Preoperative transthoracic echocardiography was performed
within 5 days before surgery and routine transesophageal assessment
was done intraoperatively in all patients. Postoperatively, all
patients were followed echocardiographically on the first postoperative
day, at the time of discharge from hospital and then 3
months afterwards. Pre- and postoperative echocardiographic
variables of all 16 patients with LA-TEA are detailed in Table 3.
Echocardiographic data were measured according to American
Society of Echocardiography criteria .
EF-ejection fraction; LVI Dd-left ventricular internal end diastolic diameter; RV Dd-right ventricular end diastolic diameter; øLA - left atrial diameter;
There was hardly any preoperative therapy before admission
to our hospital. However, substances known as ‘Traditional
Chinese Medicine’ (TCM) were taken by all of our patients. This
included different kind of teas or pastilles, but also tiger bones,
herbal and fungal ingredients and others like dried seahorses.
Although 74.2% of all patients and 100% of the 16 patients with
a LAT presented with atrial fibrillation there was no systematic
preoperative anticoagulation therapy in any. Routine preoperative
in-hospital therapy included 2x3.000 IU low molecular weight
Heparin, diuretics, Digoxin for heart rate control, and ß-blocker
as well as ACE-inhibitor medication for hypertensive patients. Ten
(8.3%) of all 120 and 3 (18.7%) of the 16 LAT patients had to be recompensated
prior to preoperative diagnostic echocardiography,
angiogram and surgery due to severe heart failure with dyspnea
at rest and problems to even lie down for the time of examination.
During the 6-month study period 26 patients have been
discharged from the hospital prior to surgery. The reasons were
refused informed consent by the patient and his relatives in 8,
referral to other hospitals for subsequent diagnosis and treatment
of suspected malignant diseases in 5, significantly pathological
and therapy resistant lung function parameters in 4 cases, and
2 patients had progressive non-infective liver failure due to
taking unknown (TCM) substances. In additional 7 patients’
surgery was refused by the combined cardiological and surgical
patient conference due to expected high perioperative risks.
In consideration of ominous threads of aggression by relatives
, one of our cardiologists was physically attacked in the cath
lab, the surgical team adapted a policy of minimizing potential
perioperative complications to prevent aggression and potential
In addition to the hospital routine anesthesia management of
patients with a preoperative systolic pulmonary artery pressure
of > 60mmHg included the control of atrial tachycardia and the
administration of phosphodiesterase type-5 inhibitor Sildenafil
(Viagra®) 25mg/p.o. at 20hrs, 10hrs and 2 hrs before surgery.
Furthermore, the introduction of a thermodilution Swan-Ganz-
Catheter, perioperative administration of nitric oxide (NO)
gas by inhalation and Milrinone lactate in combination with
Norepinephrine for weaning off extracorporeal circulation was
added to the protocol.
Surgery in all patients was performed according to the
hospital routine through a median sternotomy. Normothermic
cardiopulmonary bypass and Calafiore warm blood cardioplegia
were used for cardiac arrest . One surgeon performed all
the operations. A longitudinal right atriotomy and trans septal
left atriotomy was the conventional approach. With a massive
thrombus present, covering most of the left atrial cavity, care
was taken to search for a cleavage plane to perform a blunt
subendocardial dissection in order to separate ‘in toto’ all the thick
organized thrombotic material from the left atrial wall (Figure
4). Careful handling of the thrombotic mass was mandatory to
prevent intraoperative embolization.
The empty left atrium was carefully and thoroughly rinsed
with water and the pulmonary veins are aspirated with the
sucker to remove all possible remnants of thrombotic material. In
addition, a gauze was used to carefully clean the left atrial cavity.
Subsequent repair of rheumatic valves was performed according
to Carpentier concepts  and as described by Chotivatanapong
. Mitral valve replacement was indicated according to
American College of Cardiology/American Heart Association
Following LA-TEA and mitral repair or replacement a surgical
volume reduction of the giant enlarged left atrium was performed.
Starting the circumferential atrioplasty with a suture closure of
the left atrial appendix, the dilated tissue of the posterior and
inferior left atrial wall was plicated internally with a 5.0 Prolene
suture running clockwise toward the border of the transseptal
left atriotomy, posteriorly to the circumflex artery and along the
posterior mitral annulus (Figure 5). Care was taken to stay clear
and not involve the circumflex artery and the coronary sinus.
Patient follow-up data after discharge from the hospital were
collected at 3,6&12 months postoperatively. For the 3-month
follow-up all 16 LA-TEA patients were seen at the hospital,
a clinical assessment and a transthoracic echocardiography
was performed. At 6 & 12-months short term outcomes were
determined by direct patient contact with a telephone interview.
At 6 month a total of 3 and at 12 months additional 2 of all 16
patients could not be contacted and were lost to follow-up.
All 16 patients with LA-TEA survived surgery uneventful
without any complications. Fast track cardiac anesthesia
management allowed postoperative weaning from ventilation and
extubation between 2 and 6 hours postoperatively. Postoperative
morbidity included 2 reoperations due to a mediastinal bleeding.
There was no peri- or postoperative neurological event and
there was no 30-day mortality. The surgical technique for
thromboendarterectomy, to start with the search for a cleavage
plane to perform a blunt subendocardial dissection to remove all
thrombotic material ‘en bloc’ by LA-TEA, proved to be safe and
effective Figure 6. It should be noted that all 16 patients with LATEA
had no thrombus found in their left atrial appendix.
Preoperative blood biochemistry parameters of all 16
patients were within normal limits, except for a significantly
elevated N-terminal pro B-type natriuretic peptide (NTpBNP)
of 1,990±1,413pg/ml [normal: 100-225 pg/ml] and lactate
dehydrogenase of 263±2U/l [normal: 109-245U/l], total and
indirect bilirubin of 29.1±9.4μmol/l [normal: 2-20μmol/l]
and 19.4±6.4μmol/l [normal: 2-12μmol/l] as well as uric acid
of 430±138μmol/l [normal: 142-416μmol/l]. Preoperative
coagulation system parameters were within normal limits, except
a rather low fibrinogen of 2.2±0.4g/l [normal: 2-4g/l].
Perioperative anesthesia management required substitution
of fibrinogen and fresh frozen plasma due to fibrinogen levels
below 1.5 g/l and perioperative bleeding in 6 patients. Comparing
blood biochemistry and coagulation system parameters of the 16
LA-TEA patients with all remaining 104 of the 120 consecutive
RHD patients showed no significant differences.
All 16 patients could be discharged from the hospital
between day 10 and 15. No rehabilitation facilities or professional
postoperative care outside the hospital was available. All 16 could
be seen at the hospital for a follow up at 3 months postoperatively.
In one diabetic patient some minor local wound healing
problems required treatment. Echocardiography showed normal
functioning heart valves in all, but a smaller thrombus attached
to the wall of the LA in one patient. In 4 a minor not, significant
pericardial effusion was described and a LA function comparable
to preoperative values in all patients (Table 3).
The size of the LA was 49.3±7.8mm if compared to
61.5±6.5mm preoperatively, indicative of the LA atrioplasty
performed. All described their quality of life improved as
compared to preoperatively, asked to choose between ‘worse’,
‘same’ and ‘better’. All but 2 patients had acceptable INR values.
These 2, including the one with a small LA thrombus, did not
comply with their prescribed anticoagulation or other medication
due to financial reasons. At 6-month postoperatively one patient
was reported to have died. It was impossible to learn of any
details. Two other patients could not be found and contacted for
a telephone interview. The remaining 13 depicted their quality of
life as ‘better’ and described their physical activity comparable to
NYHA functional class I in 10 (76.9%) and class II in 3 (23.1%)
At 1 year postoperatively remaining 11 patients could be
contacted, including the one with a small LA thrombus. All
describing their quality of life as ‘better’ than before, 3 reported
to be back at work, and 4 (36.4%) characterized their NYHA
functional class as I and 7 (63.6%) as class II. Two patients
described themselves taking anticoagulation medicine only
“sometimes “, another stopped anticoagulation due to recurrent
hematoma in the leg.
Our study group represents a rare patient population that
to date might not be found in the western world. In this study
we enrolled 16 patients in rural China with massive thrombus
formation in the giant left atrium out of 120 consecutive patients
with rheumatic mitral disease. The primary findings of the study
a) That these very sick patients did astonishingly well
b) That the surgical technique for ‘en bloc’ thromboendarterectomy
proved to be safe and effective
c) That the surgical volume reduction of the giant LA with a
circumferential atrioplasty was efficient and showed positive
d) That the left atrial appendix was not the origin of LA
thrombi in these 16 mitral patients
e) That these complex procedures in very sick patients
showed to have good results after a short follow up of 1 year.
Rheumatic heart disease is an ongoing problem remaining
endemic not only in Asia but in most of the non-western world.
The total burden of cases (between 10.8 and 15.9 million in all
ages) and deaths (356000 to 524000 each year) in Asia warrants
attention and even suggests that the true numbers of RHD cases
may be even far higher than assumed . The sad story about
RHD is that it could be easily controlled and prevented. It would
be cheap and effective  and would prevent death and crippling
complications as well as economic loss and avoidable costs to the
underdeveloped healthcare systems .
However, this problem is not only neglected by media and
policy makers, it is no real focus for the WHO and ignored by
most other international organizations and receives only 0.1%
of global research and development funding [15,16]. Current
cardiovascular research focuses on minimal invasive approaches
and new percutaneous devices. All this is important for the
western world but pretty irrelevant for battling one of the top-
10 infectious causes of deaths in the world . Surgery is a
luxury not available to the poorer countries, only 11% of surgical
interventions for RHD occurred in low income countries compared
with 61% in high income countries .
There were no differences in terms of age, risk factors, valve
pathology or laboratory parameters in this group of 16 patients
who had LA-TEA if compared to all 120 RHD patients operated
during this time. There was no parameter that could be linked to
the incidence or could help to predict the risk of a LA thrombosis
in this group of 16 patients, except the rheumatic mitral valve
pathology. However, it was noted that in all of our Chinese patients’
liver specific laboratory parameters and preoperative INR have
been elevated and Fibrinogen was on the low side, but this was
only a trend and not significant.
It may be assumed that specific local dietary habits including
garlic, ginger, ginseng and others as well as herbal and fungal
TCM substances influenced these changes in blood chemistry
parameters [18,19]. Beside such specific local dietary habits
there may be some ethnic differences in the anticoagulant
activity and some genetic risk factors for the development of
venous thrombosis in the Chinese population . However, our
laboratory was not equipped for such diagnosis. N-terminal pro
B-type natriuretic peptide (NTpBNP) may be useful for diagnosing
acute decompensated heart failure and was endorsed as such a
test by international guidelines .
Furthermore, testing NTpBNP is likely to be more accurate in
Asian patients . On admission our 16 rural chinese patients
with LA thrombus had a NTpBNP plasma level of 1,990±1,413pg/
ml, considering 100-225pg/ml as normal. However, since the
group of the remaining 104 rheumatic mitral valve patients
showed a level of 1,445±960pg/ml a prevalence of significant
cardiovascular disease, ventricular dilatation, atrial fibrillation
or cor pulmonale, could not be identified for the group of 16 LA
thrombus patients. Nevertheless, these high NTpBNP plasma
levels were indicative of severe decompensated heart failure in
general for all of our patients.
On admission day 77.5% of all 120 patients were in atrial
fibrillation, but only 16 had left atrial thrombi, 6 of them had some
history of stroke and no one had any preoperative anticoagulation
or other treatment except TCM. It is often reported that the left
atrial appendix is implicated as a major location for cardiac
thrombus formation in patients with atrial fibrillation . It
seems curious, but we have not seen any thrombus formation in
the left appendix in our 16 patients. However, we found left atrial
thrombi involving extensive areas of the left atrium, sometimes
several centimeters thick.
It is a surgical challenge to remove all this thrombotic material
without leaving residual material behind and not causing atrial
perforations. Leaving residual thrombotic material behind could
be the future source of postoperative thromboembolic events.
Undetected atrial perforation will cause immediate intraoperative
bleeding complications. We found that the surgical technique of
a careful blunt subendocardial dissection in order to separate ‘in
toto’ the thick organized thrombotic material from the left atrial
wall and to perform a left atrial thromboendarterectomy seemed
to be safe and efficient. We could observe some postoperative
reduced atrial dimensions but could not detect any significant
hemodynamic effects. We did not experience any intra- or
perioperative thromboembolism or any neurological events in
this group of patients.
Previous described techniques for volume reduction of the
left atrium vary from simple to do classic plication techniques
involving the left atrial appendage [24,25] to partial auto
transplantation of the heart . Our easy-to-do circumferential
atrioplasty suture line including the left atrial appendix and
plicating the dilated tissue of the posterior and inferior wall of the
left atrium was safe and avoided extensive additional suture lines
and prolonged cardiopulmonary bypass times .
Our group of rural chinese RHD patients was very sick,
different if compared to European patients, but the prevalence of
metabolic syndrome was low. The mean body mass index of our
chinese patients was 21.6±3.2, compared to 26.3±3.9 of European
heart valve patients, the percentage of smokers 65.8% compared
to 37.2%, and the mean age was 53.9±8, but 65±14 years for
European valve patients . 75% of our RHD patients had multiple-valve surgery, 18.7% of them triple valve interventions,
but only 16.8% of European valve patients had multi valve surgery.
Operative (30 day) mortality for all our 120 rheumatic patients
was 1.7%, but 6.5% for European multiple valve patients .
Our 16 RHD patients who had left atrial thromboendarterectomy,
left atrial volume reduction and 14 of them concomitant valve
procedures, did surprisingly well. Considering that all of them
would be considered ‘inoperable’ in Europe, due to reduced
ventricular function, right heart failure, reduced lung function
and severe pulmonary artery hypertension . There was no
perioperative mortality and no neurological events. At one year
follow up there was one confirmed postoperative death, 4 were
lost to follow up, but the remaining 11 patients describing their
NYHA functional class as I and II and 3 patients even reported to
be back at work.