CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 4, May 2012
204
AFRICA
dysfunction.
14
We found that NT-proBNP and tissue Doppler
levels in patients with AF were elevated, both in the study and in
the control groups, indicating that symptomatology was not an
early marker of ventricular decompensation and that our patients
needed to be evaluated and referred for surgical intervention at
an earlier stage in the course of their illness.
The challenge in evaluating mitral regurgitation is a
determination of what really constitutes normal ventricular
function in these patients. The limitations of using the ejection
fraction in the timing of surgery become clear in subjects with
apparently normal ejection fraction and minimal symptoms.
In a study of 84 asymptomatic patients who underwent
surgical correction for MR, Agricola
et al.
15
demonstrated that
TDI systolic indices could predict postoperative left ventricular
function. In contrast, our study has shown that the TDI systolic
wave cut-off point of 0.06 m/s, which has been previously used
to rule out systolic dysfunction,
16
had very low sensitivities in
separating surgical cases from controls (see Fig. 2). Using a
higher cut-off point of 0.085 m/s only marginally improved
the specificity. It is possible that different cut-off points in
conjunction with strain measurements
17
may be more sensitive
in determining impaired LV contractile function in MR, which
can only be established in future studies with serial evaluations
at different time intervals.
Left ventricular contractile dysfunction is present in many
patients with severe MR despite a normal ejection fraction and
returns to normal after corrective mitral valve surgery in most
but not all patients.
18
This was very apparent in our patients. In
fact, all our patients experienced more than 10% decline in the
EF immediately post surgery. This was improved at the six-week
follow-up visit in all but three of our patients. Despite symptomatic
improvement, postoperative left ventricular dysfunction (EF
<
50%) has been shown to occur frequently, occurring in
close to a third of the patients successfully operated on.
19
In our study, 15 (15/27) of the patients referred for surgery
had ejection fractions above 60% and ESD values below 45 mm,
indicating that the reason for surgery in these patients was the
presence of significant symptoms while on medical treatment.
There is an inherent subjectivity in defining significant valve-
related symptoms. Indeed, when the second ROC curve was
constructed using NYHA class as the determinant, NT-proBNP
again emerged as the strongest discriminator of advanced NYHA
class.
The low specificity of 45% for NT-proBNP indicates a
high number of false positive cases, i.e. patients selected
by NT-proBNP level as requiring surgery when in fact they
were still being followed up in the clinic on medical therapy.
We believe that what is considered the high false-positive
rate with NT-proBNP actually indicates that many of these
control subjects with or without minimal symptoms actually
required surgery, rendering them true positives. Waiting for
more advanced symptoms or change in dimensions increases the
risk of left ventricular dysfunction postoperatively. Two studies,
both from the Mayo Clinic, have highlighted the poor outcomes
in patients with severe MR who were managed conservatively
rather than surgically.
20,21
In another study, Pizzaro
et al.
22
reported that NT-proBNP
level was a stronger prognostic marker than ESD or EOA and
contributed independent prognostic information additional to
other echo parameters. They showed that the BNP cut-off of 105
pg/ml (12.4 pmol/l for NT-proBNP) identified asymptomatic
patients with severe MR who were at higher risk.
Most of the subjects in our control group had values above
this level, indicating that all our patients were at risk while being
followed up at the clinic. We attributed this to the subtle nature of
symptoms, or particularly in our study, lack of awareness (on the
part of patients or patients’ caregivers) of worsening symptoms
and poor referral guidelines from peripheral healthcare centres.
This low rate of intervention has also been reported in other centres
as well.
23
The Euro Heart Survey suggests that 31% of patients
with severe valve disease and symptoms were not operated on.
24
In our study, no parameter could discriminate pre-operatively
between the four patients who exhibited persistently elevated
NT-proBNP levels at six weeks and the rest of the sample.
All four of these patients had atrial fibrillation pre-operatively,
which persisted at six weeks and at six months. There was no
indication of difficulties with myocardial preservation to suggest
this as the cause for the decline in postoperative EF by almost
30% from normal pre-operative levels.
One of the limitations of this study was that it was performed
only in patients with severe MR. The study needs to be repeated
in different subgroups with varying degrees of MR in order to
determine the time course of NT-proBNP levels in the different
stages of MR as well as in patients with ischaemic MR. Finally,
we need to determine the association of NT-pro BNP levels in a
longer follow-up study with hard end-points such as heart failure
and death.
Conclusion
We have shown that decision making regarding the timing
of surgery in this cohort of rheumatic heart disease patients
was determined largely by advanced symptomatology, so that
patients are referred to surgery late in the course of severe MR,
with a risk of permanent LV decompensation. We propose that
NT-proBNP level be an additional marker, particularly in less
symptomatic patients, even if the EF is normal. In time, it may
prove to be the composite marker for the assessment of LV
decompensation. These findings support those of Detaint
et
al.
25
in that the BNP level reflects the severe haemodymamic,
ventricular and atrial consequences of MR.
References
1.
Griffin BP. Timing of surgical intervention in chronic mitral regurgita-
tion: is vigilance enough?
Circulation
2006;
113
(18): 2169–2172.
2.
Rosenhek R, Rader F, Klaar U, Gabriel H, Krejc M, Kalbeck D,
et al
.
Outcome of watchful waiting in asymptomatic severe mitral regurgita-
tion.
Circulation
2006;
113
: 2238–2244.
3.
Weber M, Dill T, Arnold R,
et al
. N-terminal B-type natriuretic peptide
predicts extent of coronary artery disease and ischemia in patients with
stable angina pectoris.
Am Heart J
2004;
148
: 612–620.
4.
Zoghbi WA, Enriquez-Sarano M, Fosrer E,
et al.
American Society of
Echocardiography. Recommendations for evaluation of the severity of
native valvular regurgitation with two-dimensional and Doppler echo-
cardiography.
J Am Soc Echocardiogr
2003;
16
: 777–802.
5.
Thomas L, Foster E, Hoffman JIE, Schiller NB. The mitral regurgita-
tion index: an echocardiographic guide to severity.
J Am Coll Cardiol
1999;
3320
: 16–2022.
6.
Januzzi JL, Carlos AC, Saif A,
et al.
The N-terminal pro-BNP
Investigation of Dyspnoea in the Emergency Department (PRIDE)
Study.
Am J Cardiol
2005;
95
: 948–954.
7.
Cowie MR, Jourdain P, Maisel J, Dahlstrom
U. Clinical applica-
tions of B-type natriuretic peptide testing.
Eur Heart J
2003;
24
(19):
