Left Ventricular Diastolic Dysfunction in COPD May Manifest Myocardial Ischemia

Abstract

Background: Left ventricular diastolic dysfunction (LVDD) is observed frequently in advanced COPD and it appears unrelated to the co-presence of its common risk factors as hypertension, diabetes, ischemic heart disease, etc. The patho-physiology of this association is yet unclear. We postulated that ischemia not apparent in the routine clinical evaluation could be the cause of such LVDD. Method: Advanced COPD (GOLD III & IV) patients being excluded of concomitant presence of hypertension, diabetes, ischemic heart disease, and hypothyroidism by conventional evaluations were looked for the presence of LVDD by tissue Doppler echocardiography. The subjects having LVDD were further evaluated with 99mTc MIBI stress myocardial perfusion imaging to detect the presence of ischemia. Results: 7out of 14 patients of advanced COPD sufferers without any known risk factors for LVDD had reversible perfusion defect in the stress perfusion imaging predominantly involving the inferior wall myocardium (71.43%). Most of the defects, though significant, were not picked up by stress ECG. Conclusion: Reversible ischemic defects are quite common (50%) in advanced COPD patients with LVDD without the presence of common risk factors. This association needs further evaluation.

Keywords: :

• COPD
• Echocardiography
• diastolic dysfunction.

Abbreviations
COPD Chronic Obstructive Pulmonary Diseases	=
ECG Electrocardiogram	=
GOLD Global initiative for Chronic Obstructive Lung Disease	=
LV Left ventricle	=
LVDD Left Ventricular Diastolic Dysfunction	=
99m Tc MIBI 99 meta stable Technitium Methoxy isobutyl isonitrile	=
RV Right Ventricle	=
SPECT Single Photon Emission Computed Tomography	=

Introduction

COPD can affect the functions of both the right and the left heart in different ways. The changes in the right ventricular function secondary to pulmonary artery hypertension is well known, but the changes in left ventricular functions in COPD is yet elusive and presently an area of interest for researchers. Patients of COPD having no known cardiac problem show left ventricular contractile (systolic) dysfunction in as low as 5% (1). However, studies have recorded significantly high incidence of relaxation abnormalities (diastolic dysfunction) especially in the severe COPD populations (2–4) and this appears unrelated to the presence of pulmonary hypertension (5). The presence of LVDD in COPD again appears to be independent of the common known risk factors as hypertension, diabetes and ischemic heart disease (6, 7). Since the knowledge available in this area is far from adequate and the pathogenesis of such LVDD is unclear, we hypothesized that ischemia not apparent in ECG and resting echocardiography done in the prevailing clinical practice could be the cause. To substantiate, we have evaluated a cohort of patients with advanced COPD and LVDD (without any risk factors) for evidence of myocardial ischemia with stress isotope myocardial perfusion imaging.

Objective

To determine whether ischemia not apparent in resting ECG and echocardiogram could be the cause of diastolic dysfunction in advanced COPD patients having no risk factors for left ventricular relaxation abnormalities.

Materials and Methods

This work has been approved by the Ethics Committee of the Institute. The patients who took part in this study had given their consent accordingly. COPD patients (age 45–70 yrs), confirmed by Spirometry according to ATS guideline (8) were randomly selected and subsequently staged as per the GOLD guideline (9). Subjects with a presence of arrhythmia or showing left ventricular systolic dysfunction and regional wall motion abnormality in resting echocardiography or evidence of valvular or congenital heart disease (if echocardiography was being done earlier), or cardiomyopathy (proved or suspected) were excluded. Any patient with history of lower respiratory tract infection within the last (preceding) 1 month, or having evidence of active and concomitant pulmonary disease other than COPD or any significant renal, hepatic or other systemic problem were also excluded along with patients with baseline resting SpO₂ <90%.

The willing subjects of COPD (GOLD III/IV) were kept for a 2-week run in period on inhaled drugs only after signing a written informed consent. During this period, the patients were evaluated for known risk factors of diastolic dysfunction as hypertension, diabetes mellitus, ischemic heart disease, and hypothyroidism observing the following criteria:

1. Hypertension was labeled when a patient had recorded history of hypertension of stage one and above (systolic pressure ≥140 mm of Hg and diastolic ≥ 90 mm of Hg) or has been on any antihypertensive drug (10);

2. Diabetes mellitus was diagnosed in patients with history of diabetes and treatment with antidiabetic drugs or having recorded fasting plasma sugar of ≥126 mg/dl and casual plasma glucose concentration ≥ 200 mg/dl with or without typical diabetic symptoms, (11);

3. Ischemic heart disease was defined when a patients had history of anginal chest pain and suggestive changes in 12 lead ECG or has been on anti-ischemic drugs with previously performed ECG showing positive changes. Moreover, any regional wall motion abnormality in echocardiographic evaluations was regarded as evidence of the presence of ischemic heart disease; and,

4. Hypothyroidism was excluded on the basis of elevated TSH with a normal or low free T4 and T3.

The patients having none of the above mentioned risk factors were then evaluated by a single dedicated cardiologist doing echocardiography (being blinded about the study) by using 2D Doppler and Tissue Doppler echocardiography for the presence of diastolic dysfunction. The tests were done in the morning hours only on HP sono-7005h (Aligent Technology, USA) machine.

The diagnosis of LVDD was based on positivity of any 3 of the 5 following criteria by Pulse-Wave Doppler and Tissue-Doppler examination:

1. E/A velocity ratio of less than 1 by PW at tips of mitral valve leaflets;

2. IVRT (more than 90 m sec) estimated by CW Doppler kept between LV outflow and inflow in apical 5 chamber view;

3. Deceleration time of E wave of mitral flow by PW Doppler to be less than 150 m secs;

4. Atrial filling period more than 20 m seconds;

5. Ratio of E/E’: {More than 10}; and,

6. E’ velocity by Tissue Doppler at Mitral annulus as less than 8 cm/sec.

Thereafter, those patients having LVDD underwent stress SPECT myocardial perfusion imaging with 99mTc-MIBI (by INFINIA II (GE) using Dual Head Gumma Camera) to look for the presence of ischemia. All patients were studied in the morning hours with a prefixed uniform preparation. The patients were subjected to dynamic exercise on Treadmill following the BRUCE protocol in majority of the patients. In a few patients with poor exercise tolerance due to arthritis, obesity, old age etc. were tested by modified Bruce protocol. The exercise was terminated with development of discomfort and or fatigue or achievement of the target heart rate. 99mTc MIBI (8 to 10 mCi) was injected at peak exercise and tomographic images were obtained after 1 hour and again at 3 hours following the injection of a higher dose of 20–30 mCi for perfusion image at resting phase. The images were reconstructed in short axis, vertical long axis and horizontal long axis and the perfusion defects, if any, were looked for the site, extent (small, moderate, or large), severity, and the reversibility as per standard protocol. Quantifications of the defects were calculated by standard software GE Myovation with Quantitative Gated and Perfusion SPECT.

Results

A total of 77 COPD patients were screened. The mean age was 62.89 (±8.40) years and the mean FEV1% was 29.75 (±13.69) liters and mean SaO₂ of 95.70 (±3.00)%. Out of these patients, 11 were excluded for poor performance of Spirometry, 9 patients lost to follow-up, and 19 patients were excluded for having one or more risk factors for LVDD. The risk factors present were hypertension (68.18%), hypothyroidism (22.72%) and diabetes (9.09%). Out of these 38 patients having no coexistent risk factors for LVDD, 27 underwent echocardiography. Evidence of diastolic dysfunction was detected in 16 patients (59.26%) and none of them had dilated or hypertrophied right ventricle with mean RV/LV ratio of 0.39 suggesting no significant RV pressure overload. There were no obvious differences in general, lung function and echo-cardiographic parameters between the patients with or without LVDD (Table 1). 14 out of these 16 patients were subjected to SPECT-MIBI perfusion study; 7(50%) had evidence of reversible perfusion defect predominantly affecting the inferior wall myocardium (71.43%).

Table 1.  The general, lung function, and echocardiographic characteristics of COPD population with or without left ventricular diastolic dysfunction

Parameters	COPD with DD Mean ± SD (n = 16)	COPD without DD Mean ± SD (n = 11)	p-value with 95% CI
Age (years)	64.31 ± 10.70	61.09 ± 5.56	NS (−4.04, 10.48)
BMI	20.65 ± 3.80	19.16 ± 3.27	NS (−1.41, 4.39)
FEV1 (%)	25.68 ± 6.91	24.10 ± 10.77	NS (−5.41, 8.57)
Resting heart rate (/min)	94.69 ± 11.34	97.00 ± 14.42	NS (−7.90, 12.52)
E/A ratio	0.72 ± 0.18	1.09 ± 0.33	<0.001* (0.17, 0.57)
IVRT	113.00 ± 2.94	83.82 ± 19.14	<0.001* (19.24, 39.12)
D-time	143.31 ± 12.12	169.91 ± 10.97	<0.001* (17.18, 36.02)
E’ (cm/sec)	8.68 ± 1.82	11.78 ± 2.32	<0.001* (1.46, 4.74)
E/E’	8.27 ± 3.08	6.77 ± 2.12	NS (−0.71, 3.71)

Criteria entertained for presence of diastolic dysfunction: when any 3 out of the 5 of the following parameters are present - E/A less than 1; IVRT more than 90 msecs; D-time less than 150 msecs, E/ E′ more than 10; E′ less than 10.

NS = not significant; *statistically significant.

Discussion

The LV diastolic dysfunction in COPD patients has long been a neglected area. This is possibly because most of the COPD victims are elderly with commonly happening different co-morbidities that can lead to impaired diastolic relaxation. Of late, people have shown interest in probing this particular entity and several reports of LVDD with or without hemodynamic data being published (5); with the prevalence of the condition being noted as quite high in advanced COPD patients. The present study also tallies to this showing a high frequency (59.26%) of apparently unexplained LVDD in advanced COPD subjects (6, 7). The compounding effect of age is unlikely because the difference of age between COPD patients with or without LVDD is not obvious (Table 1).

Several mechanisms of diastolic dysfunction in COPD are described in literature as, a) right ventricular dilation and displacement of the interventricular septum towards left ventricular lumen hindering normal LV filling (12,13), (b) hypoxia (5,14), (c) prolonged use of beta 2 agonist (15) and d) pericardial restriction and impedance generated by hyper-inflated lungs (16). A recent study has shown that raised PAP is not related to the presence of LVDD in COPD (5) and thus negating the most prevalent theory of raised RV pressure induced LV filling defect. This leads to a room for further hypothesis regarding the mechanism of LVDD in COPD. It is possible that factor/factors is/are not picked up in usually performed investigations in clinical practice could play a role here. Dealing with the COPD subjects for years with our interest being preserved for the left ventricular diastolic dysfunction, we thought that the missing link could be myocardial ischemia not apparent in routine ECG and echocardiography. Thus, we decided to evaluate these patients with stress myocardial perfusion scintigraphy.

Myocardial ischemia and coronary artery disease (CAD) could be co-related but these two terms are not synonymous. COPD, being a systemic inflammatory disease, has been postulated and demonstrated to predispose the development of CAD (17, 18). However, by virtue of the presence of hypoxemia there could be development of collateral circulation mediated by release of VEGF. This may actually protect the patients from myocardial ischemia (19). Thus, we postulated that the manifestation of myocardial ischemia in COPD could be a different phenomenon and that the COPD –myocardial ischemia relationship may be a dynamic and far more complicated as being perceived by us (7). This led us to hypothesize that the observed LVDD could be a result of myocardial ischemia not apparent in routine clinical evaluations.

Our view has been supported by a recent publication that has changed the concept of the causes of COPD mortality. So far, cardiovascular diseases have been thought to be the leading cause of hospitalization and second leading cause of death in mild to moderate COPD (20, 21, 22) while respiratory failure has been implicated as the most important cause of death in advanced (GOLD III and IV) disease (22, 23). TORCH (Towards a Revolution in COPD Health) study also support this view that only 10% patients of COPD (GOLD III and IV) died of a clear cut cardiovascular event such as stroke, myocardial ischemia and heart failure (24). It is difficult to accept and explain such findings, since advanced COPD patients are survivor of mild to moderate disease and if cardiovascular events predominate in early stage of the disease, how come the patients of advanced stage get spared of them and how does the cardiovascular manifestations (predominantly ischemic) escape the attention of the clinicians with routine evaluations that should have detected the changes mentioned.

Accepting the fact that COPD is not a homogeneous disease, the plausible explanations would be a) that with the advancement of the disease and development of the ongoing hypoxemia, there is development some mechanism like myocardial vascular collateralization that prevents and perhaps overcorrects the ischemic defect out of coronary artery disease (CAD), or b) that the ischemia in severe/advanced COPD subjects present differently from the conventional ways to escape detection. In advanced COPD subjects, the dyspnoea and wheezing are more prominent symptoms and it is possible that the attention of the patient as well as the clinicians gets shifted away from any possible manifestations of ischemia. Therefore, on rational thinking, the second possibility looks more acceptable, and, here, a the recent information has indirectly strengthened our view by unraveling the cause of death in COPD exacerbation through a series of consecutive autopsies on patients who died within 24 h of hospitalization with diagnosis of acute exacerbation of COAD (25). It showed that cardiovascular mortality, not the respiratory failure happens to be the leading cause of death in hospitalized patients with COPD exacerbations contrary to the prevailing concept (25). Another important and accompanying revelation was that the routinely collected clinical data could not predict the cause of death in any of these patients before the terminal event (22).

We thought that LV DD could be the missing link to detect ischemia. So we had chosen stress myocardial scintigraphy as an alternative modality to assess the presence of ischemia in advanced COPD patients. The results reveal that 59.26% of GOLD III and IV COPD subjects have isolated LVDD not compounded by any commonly known risk factors (6). Fifty percent of these patients with unexplained LVDD got reversible myocardial perfusion defect signifying the definite presence of ischemia (7).

Interestingly, the data also reveals that a) the ischemic defects in stress MIBI scan were not picked up in stress ECG, b) the defects are mostly moderate or significant and c) they are mostly confined to inferior wall alone or with other walls of myocardium. In view of the fact that the sensitivity of stress nuclear imaging is higher than that of Stress ECG (88% vs. 68%) (25), the present findings of such a huge difference in positivity of stress nuclear scanning compared to the stress ECG appears very unlikely for a co-incidence (though the number of patients being small). Whether use of β₂-agonist, xanthines or changes in pH, electrolytes in COPD can decrease the sensitivity of stress ECG is a matter of conjecture. It is again a topic of investigation why the inferior wall of the myocardium is affected more in COPD from ischemia and how such a right coronary artery disease (that commonly supplies the inferior wall of the heart) can affect the relaxation property of the left ventricle (that is mostly supplied by the left coronary artery).

We postulate that the sensitivity of detection of the inferior wall ischemia by 12-lead ECG might be poor as the same is detected by only 3 leads (II, III, and aVF) whereas the apical, anterior or antero-lateral walls are represented by more number of ECG leads. Again, the imaging of inferior wall by 2D Echo in patients of COPD is difficult because of poor echo window in these patients. Thus, unlike patients with anterior/antero-lateral wall ischemia, inferior wall ischemia patients could not be excluded from study because of poor sensitivity of the screening tests (ECG and RWMA by echocardiography). We wish to put forward yet another explanation. As the right coronary artery, in most of the persons, supply the inferior wall of LV and the same artery supplies the RV through a proximal branch, hypertrophy of RV as well as higher afterload in RV due to cor-pulmonale can produce a steal-phenomenon from inferior wall territory. Such postulated relationship of myocardial perfusion dynamics demands in-depth evaluations.

Diagnosis of diastolic dysfunction by transthoracic echocardiography may be difficult. So we included multiple parameters including TDI to rule out any chance of misinterpretation from pseudonormalization of blood flow and motion relations of myocardium. The patients with or without diastolic dysfunction had clear-cut difference in all parameters concerned.

Noninvasive diagnosis of the presence of ischemia can be done effectively with 99mTcMIBI myocardial perfusion imaging. We looked for presence of reversible perfusion defect on the SPECT image on a defined protocol. The result has been an interesting and eye opening one: out of 14 patients 7 (50%) had a reversible ischemic change. It is worthwhile to notice that the changes are significant in their extent with 50% being of moderate to severe in intensity.

Finally, it may be prudent to say that the conventional evaluation of COPD does not give any clue to a cardiovascular morbidity probably because the cardiovascular signs including that for ischemia may be different in advanced COPD patients. Thus, the presence of apparently unexplained diastolic dysfunction could be the missing link to suspect ischemia in advanced COPD patients. Furthermore, it will be interesting to see the stress myocardial perfusion pattern in COPD subjects with diastolic dysfunction of LV having co-morbid states that are not related to myocardial ischemia as hypertension, diabetes etc.

There are some serious weaknesses of the study. First, the number of patients undergoing stress SPECT MIBI study is small. Second, there should have been a age and sex matched control group of subjects who have no risk factor for LVDD and no COPD. Third, a third group patients with COPD and no diastolic dysfunction should have been incorporated for comparison.

Conclusions

Diastolic dysfunction in COPD patients without the common coexistent risk factors has got significant association with myocardial ischemia not apparent in conventional evaluations. This association demands further in-depth studies and detailed investigations. This will certainly help us in treating COPD patients better in future.

Declaration of Interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. The study was done with financial support from the Department of Science and Technology, Government of West Bengal, India.