2.5 Statistical analysis:
Continuous data are presented as median (range) or mean (standard
deviation) after testing for normality with the Shapiro-Wilk test;
categorical data are presented as frequencies and percentage. Mean
comparison was performed using Student’s t test or Wilcoxon Mann-Whitney
test as appropriate. Proportions were compared using Chi-square test or
exact methods as necessary. Pearson and Spearman coefficients were used
to assess correlations between continuous data. A receiver operator
curve (ROC) analysis was used to determine the diagnostic accuracy of
NTproBNP for LVMD. The best cut-off value of NT-proBNP to detect LVMD
was empirically estimated based on the Liu method, and values of
sensitivity (Se), specificity (Sp), negative predictive value (NPV) and
positive predictive value (PPV) were calculated for the obtained cut
point. Prediction models were developed using multivariate logistic
regression analysis to identify potential predictors of PICU admission.
The selection of the included variables was based on the theoretical
background and the exploratory analysis. The discriminating ability of
each model was assessed by the area under the receiver operator
characteristic (AUC) curve. The AUCs from the obtained models were then
compared by using the DeLong method[34] to determine whether any
model resulted in increased predictive accuracy. The reliability of
echocardiographic measurements was evaluated with the intra-class
correlation (ICC) coefficients and Bland-Altman (BA) analysis[35].
Based on the 95% confident interval of the ICC coefficients, values
less than 0.5, between 0.5 and 0.75, between 0.75 and 0.9, and greater
than 0.90 were considered indicative of poor, moderate, good, and
excellent reliability, respectively. All the statistical analyses were
performed using the Stata 13.0. (StataCorp. 2013. Stata Statistical
Software: Release 13. College Station, TX: StataCorp LP.). A P value
< 0.05 was considered statistically significant.
3. Results
3.1. Baseline characteristics and outcomes of patients
with RSVB: We enrolled a total of 50 cases of RSVB with a median age of
2 (1-6.5) months (40% female). The control group consisted of 50
healthy infant with no differences regarding age, sex or BSA
distribution (Table 1) . RSVB patients were admitted 2.76 (1.23)
days after the initial symptoms, with a median BROSJOD score of 6
(1-14), a median SpO2 of 93% (87-98%) , and a median heart rate of 118
bpm (89-179 bpm). Up to 9 (18%) cases presented RA and a BROSJOD score
> 10. A total of 10 (20%) cases needed PICU admission
within 1.20 (0.38) days from hospitalization (length of PICU stay 5
(2-9) days), and were classified as having a severe RSVB. 3 of these10
cases required MV and 7 of them required CPAP/BiPAP. No cases of
arrhythmia different from sinus tachycardia were observed. No patient
required inotropic support and none of the included patients died.
3.2. Echocardiographic alterations in patients with
RSVB: The RSVB group had a higher proportion of pericardial effusion
(34% vs 6%; p<0.001). All cases of pericardial effusion were
mild and did not require any treatment. We observed higher values of
LVTX (0.42 vs 0.36; p=0.008) in infants with RSVB than in controls(Figure 3). The RSVB group also presented more cases of septal
flattening (28% vs 6%; p=0.003), and higher TRJG (27 vs 22 mmHg;
p=0.013), RVTX (0.39 vs 0.36; p=0.005) and LVEIs (1.08 vs 1;
p<0.001) than the control group. There were no differences
between RSVB and control groups regarding echocardiographic parameters
of ventricular dimensions, systolic or diastolic function (Table 1).
3.3. Adverse LV-RV interactions in patients with RSVB:In the RSVB group, increased LVTX was related to echocardiographic
parameters indicating higher RV dimensions (Rho RVDD = 0.56, Rho
RVLVr = 0.60), higher RV pressures (Rho TRJG = 0.54, RhoATET = -0.50, Rho LVEI = 0.77), and decreased RV global function
(Rho RVTX= 0.74). We did not find associations of
echocardiographic measurements of RV systolic function (TAPSE and St)
with LVTX (Table 2 & Figure 4).
3.4. LVMD in patients with RSVB: We found LVMD, defined by a
LVTX > 0.5, in 9 (18%) patients with RSVB (Table
3). LVMD was associated to PICU admission (89% vs 5%;
p<0.001). These patients presented higher BROSJOD scores (11
vs 6; p<0.001), lower SpO2 (90% vs 94%; p <0.001),
more cases of RA (55% vs 9%; p = 0.001), and higher NT-proBNP levels
(2221 pg/ml vs 377 pg ml; p<0.001) than those with a normal LV
myocardial function. The LVTX was also strongly correlated with
NT-proBNP levels (Rho = 0.73) (Figure 5).
3.5. NTproBNP as biomarker for LVMD: The diagnostic performance
of NTproBNP to predict LVMD in infants with RSVB resulted high with an
area under the ROC curve of 0.91 (CI95% 0.79-0.98) (Figure 6).The best estimated cut-off value to predict LV myocardial dysfunction on
echocardiography was 1500 pg/ml, that correctly classified 92% of
cases, and presented a Se of 0.80 (CI95% 0.49-0.94), Sp of 0.95 (CI95%
0.83-0.98), a PPV of 0.80 (CI95% 0.49-0.94), and a NPV of 0.95 (CI95%
0.87-0.99) (Youden index 0.75).
3.6. Clinical and laboratory predictors of PICU
admission in RSVB: We developed 3 different predefined prediction
models for PICU admission in RSVB. The variables used were clinical and
laboratory parameters that are usually used to assess severity in RSVB
(age, clinical score and RA), and NT-proBNP as biomarker for LVMD. Model
1 included age < 3 months, BROSJOD score > 10,
RA, and NTproBNP > 1500 pg/ml. Model 2 included age
< 3 months, BROSJOD score > 10, and RA. The model
3 included age < 3 months and BROSJOD score > 10.
The diagnostic yield of model 1 for PICU admission was excellent (AUC
0.945, CI95% 0.880-1), and significantly higher than the yields for
model 2 (p= 0.033) and model 3 (p=0.026). There were no differences
between ROC curves of models 2 and 3 (p=0.432). (Table 4 &
Figure 7). In model 1, the presence of NT-proBNP levels >
1500 pg/ml was the only independent predictive factor for PICU admission
in RSVB, with an OR 27.03 (CI95% (1.50-487), p=0.025).
4. Discussion
The main finding of our study is that LVMD was observed at early stages
of the disease in up to 18% of previously healthy infants with RSVB
when assessed by DTI-derived LVTX. The LVMD was associated with a worse
respiratory state, PICU admission, and echocardiographic signs of RV
pressure overload and RVMD, indicating the presence of adverse RV-LV
interactions in cases of severe RSVB. Also, we observed that NTproBNP
identified accurately LVMD. Moreover, we found an added benefit to the
addition of NTproBNP to the clinical evaluation in predicting the
development of a severe disease in this population.
CHD are important causes of morbidity and mortality in RSVB[5]. This
may be related to multiple physiological factors including baseline
compromised cardiorespiratory function and the potential development of
PH. However, little is known about LVMD and its association with RV
function, pulmonary hemodynamics, and outcomes in previously healthy
infants with RSVB. In accordance with literature, the present study
reveals that conventional parameters of myocardial function are not
altered in RSVB [17-19]. Only one previous study has assessed
myocardial performance using the TEI index in RSVB [12]. Our results
are consistent with those reported by Thorburn et al. who found a RVMD
in ventilated patients with severe RSVB. However, they did not
demonstrate PH as the cause of RVMD. This could be due to the use of
only TRJG as echocardiographic marker of RV pressures, and the presence
of PH might had been underestimated. In a recent work from our group we
used a combination of different echocardiographic parameters to assess
RV pressures, PH was reported in up to 22% of RSVB cases at early
stages of the disease, but accordingly with Thorburn et al, we did not
found associations between PH and RV or LVMD [13]. Remarkably, we
assessed the ventricular function only by conventional parameters (TAPSE
and LVSF).
To the best of our knowledge, this is the first study to evaluate LVMD
using TDI-derived LVTX, and to assess RV-LV interactions in infants with
RSVB. LVTX, which includes both systolic and diastolic time intervals to
assess the global cardiac dysfunction, is an easily performable,
recordable and reproducible parameter with normal reference values that
can be applied to the entire spectrum of the paediatric population, with
no impact of age, heart rate, and BSA[36]. Using LVTX we found a
LVMD in near 20% of cases. The RV shares muscle fibers, the
inter-ventricular septum (IVS), and the pericardial sac with the LV.
Consequently, changes in RV affect the LV, a concept termed ventricular
interdependence [16,37,38]. In this study, we observed a moderate to
strong correlation between LVTX and leftward displacement of the IVS
(LVEI), raised RV pressures (TRGJ, ATET) and reduced RV global function
(RVTX), supporting our hypothesis that through ventricular
interdependence, RV pressure overload and RV dysfunction could induce
LVMD in RSVB. These observations could add new insights into the
pathophysiology of RSVB, highlighting a key role of the cardiovascular
system, especially LV myocardial performance in this setting. Should our
findings be validated in other similar populations in different settings
be, new therapeutic goals and approaches based could be explored
(instauration of early respiratory support in cases with increased
NT-proBNP, pulmonary vasodilators to reduce RV pressure overload,
avoiding epinephrine in cases with LVMD), in a disease without any
current effective treatment.
In RSVB, prevention and early identification of infants at risk for
severe disease in order to provide the best management options and
potentially decrease morbidity are the major goals. Current guidelines
recommend only clinical observation for this purpose in infants without
known comorbidities [3]. However, most clinical scores for RSVB are
not well validated and fail to predict outcomes [39,40]. Recently,
the BROSJOD score, a validated clinical score for RSVB, has shown a
strong capacity to predict the evolution in RSVB, but it is not
generalizable due to the single-centre character of the study [26].
In this context, the identification of novel biomarkers with adequate
predictive value for disease severity in RSVB is an area of increasing
research interest. Neutrophins, cytokynes and leukotrienes are promising
but not widely available for clinical practice [41]. Also, previous
studies have tested cardiac troponin as prognostic marker but with
inconsistent results [9,10,12,13]. The LVMD found in our population
was only identifiable by TDI-echocardiography, suggesting that it was
mild in most of our patients. Nevertheless, it does not mean that LVMD
is inconsequential in RSVB. Remarkably, most patients with LVMD
presented a severe disease at admission and most of them required PICU
admission. Of note, we included non-ventilated infants at early stages
of the disease (mean time of 2.76 (1.23) days after the initial
symptoms), when the patients have not been admitted yet to PICU,
increasing the prognostic value of our results. Therefore, assessing and
understanding myocardial function in RSVB seems to be relevant.
Another interesting finding of our study was that NT-proBNP could be a
useful biomarker for LVMD and subsequently outcomes in RSVB. Previous
studies have also documented the correlation between LVTX and NT-proBNP
plasmatic levels [42-44]. We explored the diagnostic accuracy of
NT-proBNP to detect LVTX > 0.50, which resulted high (AUC
0.91), with an optimal cut-off value of 1500 pg/ml (Se 0.80, Sp 0.95,
PPV 0.80, NPV 0.95). We also tested the benefit to add NTproBNP to the
currently used clinical and laboratory data to assess outcomes in RSVB.
Although the predictive models including age, BROSJOD score and RA
presented a high predictive accuracy for PICU admission in our
population, we observed that the addition of NTproBNP to the models
increased significantly their predictive value, and that NTproBNP was
the only independent factor within the analysed that predicted a severe
course of the disease. Most cases of RSVB are mild forms manageable on
an outpatient basis without need of laboratory exams. Nevertheless, many
children sufficiently ill to require hospitalization will routinely have
laboratory studies drawn. Adding NTproBNP determination to these studies
could be useful, in order to identify high-risk patients that benefit of
echocardiographic screening to PH or LVMD. Based on our results it might
be reasonable to perform an echocardiogram in those patients with
NT-proBNP levels > 1500 pg/ml.
This work included some limitations. Our study was performed at a single
center and with a relatively small size. We excluded irritable or
unstable patients, where the technical difficulties due to respiratory
comorbidity and the patient’s worse tolerance to the evaluation could
impact the results. The PICU admission criteria, and subsequently the
definition of severe diseases in the study, are based on the protocol of
our hospital, which can vary between different institutions. Therefore,
a larger multicentre cohort study including irritable or unstable cases
and with uniform PICU admission criteria may be needed for verification
and generalization of our results. Finally the assessment of myocardial
function and PH on echocardiography was not confirmed by an independent
gold-standard method, such as cardiac magnetic resonance imaging or
right heart catheterization, and therefore some patients could have been
miss-classified.
5. Conclusions
LVMD secondary to adverse RV-LV interactions could be present in healthy
infants with RSVB during early stages of the disease, negatively
impacting the outcome. NTproBNP seems to be an adequate biomarker for
LVMD and adding NTproBNP to current recommendations to assess outcomes
in RSVB could help to detect early those cases that develop a severe
illness. Future research is need in order to confirm these results and
to design new therapeutic approach based on them.
Author Contributions: C onceptualization, RG.M; methodology,
RG.M & BF.I.; software, RG.M & BF.I; validation, RG.M., BF.I., PR.A.,
CM.A., LL.S.; formal analysis, RG.M & BF.I.; investigation, RG.M.,
PR.A. and CM.A..; resources, RG.M & LL.S.; data curation, RG.M, CM.A.
and PR.A.; writing—original draft preparation, RG.M.; writing—review
and editing, RG.M., BF.I., PR.A., CM.A:, LL.S..; visualization, RG-M. &
BF.I..; supervision, RG.M & BF.I..; project administration, RG.M.
Funding: This research received no external funding.
Conflicts of Interest: The authors declare no conflict of
interest.
References
1. Hasegawa K, Tsugawa Y, Brown DFM, Mansbach JM, Camargo CA. Trends in
bronchiolitis hospitalizations in the United States, 2000-2009.
Pediatrics. American Academy of Pediatrics; 2013 Jul;132(1):28–36.
2. Ghazaly M, Nadel S. Characteristics of children admitted to intensive
care with acute bronchiolitis. Eur J Pediatr. 2018 Jun;177(6):913–20.
3. Ralston SL, Lieberthal AS, Meissner HC, Alverson BK, Baley JE,
Gadomski AM, et al. Clinical practice guideline: the diagnosis,
management, and prevention of bronchiolitis. Vol. 134, Pediatrics.
American Academy of Pediatrics; 2014. pp. e1474–502.
4. Willson DF, Landrigan CP, Horn SD, Smout RJ. Complications in infants
hospitalized for bronchiolitis or respiratory syncytial virus pneumonia.
J Pediatr. 2003 Nov;143(5 Suppl):S142–9.
5. Tulloh RMR, Medrano-Lopez C, Checchia PA, Stapper C, Sumitomo N,
Gorenflo M, et al. CHD and respiratory syncytial virus: global expert
exchange recommendations. Cardiol Young. 2017 Oct;27(8):1504–21.
6. Fitzgerald D, Davis GM, Rohlicek C, Gottesman R. Quantifying
pulmonary hypertension in ventilated infants with bronchiolitis: a pilot
study. J Paediatr Child Health. 2001 Feb;37(1):64–6.
7. Bardi-Peti L, Ciofu EP. [Bronchiolitis and pulmonary
hypertension]. Pneumologia. 2010 Apr;59(2):95–100.
8. Kimura D, McNamara IF, Wang J, Fowke JH, West AN, Philip R. Pulmonary
hypertension during respiratory syncytial virus bronchiolitis: a risk
factor for severity of illness. Cardiol Young. 2019 May 20;20:1–5.
9. Clark SJ, Eisenhut M, Sidaras D, Hancock SW, Newland P, Thorburn K.
Myocardial injury in infants ventilated on the paediatric intensive care
unit: a case control study. Crit Care. BioMed Central; 2006;10(5):R128.
10. Moynihan JA, Brown L, Sehra R, Checchia PA. Cardiac troponin I as a
predictor of respiratory failure in children hospitalized with
respiratory syncytial virus (RSV) infections: a pilot study. The
American Journal of Emergency Medicine. 2003 Oct;21(6):479–82.
11. Thorburn K. Cardiac Troponin T levels and myocardial involvement in
children with severe respiratory syncytial virus lung disease. Acta
Paediatr. 2004 Jul;93(7):887–90.
12. Thorburn K, Eisenhut M, Shauq A, Narayanswamy S, Burgess M. Right
ventricular function in children with severe respiratory syncytial virus
(RSV) bronchiolitis. Minerva Anestesiol. 2011 Jan;77(1):46–53.
13. Rodriguez-Gonzalez M, Benavente Fernández I, Castellano-Martinez A,
Lechuga-Sancho AM, Lubián-López SP. NT-proBNP plasma levels as
biomarkers for pulmonary hypertension in healthy infants with
respiratory syncytial virus infection. Biomarkers Med. 2019 Jun 3.
14. Driessen MMP, Meijboom FJ. Adverse ventricular-ventricular
interactions in right ventricular pressure load: Insights from pediatric
pulmonary hypertension versus pulmonary stenosis. Physiol Rep. 2016
Jun;4(11):e12833.
15. Burkett DA, Slorach C, Patel SS, Redington AN, Ivy DD, Mertens L, et
al. Impact of Pulmonary Hemodynamics and Ventricular Interdependence on
Left Ventricular Diastolic Function in Children With Pulmonary
Hypertension. Circ Cardiovasc Imaging. 2016 Sep;9(9):e004612.
16. Motoji Y, Tanaka H, Fukuda Y, Sano H, Ryo K, Imanishi J, et al.
Interdependence of right ventricular systolic function and left
ventricular filling and its association with outcome for patients with
pulmonary hypertension. Int J Cardiovasc Imaging. Springer Netherlands;
2015 Jan 23;31(4):691–8.
17. Pahl E, Gidding SS. Echocardiographic assessment of cardiac function
during respiratory syncytial virus infection. Pediatrics. 1988
Jun;81(6):830–4.
18. Esposito S, Salice P, Bosis S, Ghiglia S, Tremolati E, Tagliabue C,
et al. Altered cardiac rhythm in infants with bronchiolitis and
respiratory syncytial virus infection. BMC Infect Dis. 2nd ed. BioMed
Central; 2010 Oct 24;10(1):305.
19. Horter T, Nakstad B, Ashtari O, Solevåg AL. Right and left
ventricular function in hospitalized children with respiratory syncytial
virus infection. Infect Drug Resist. Dove Press; 2017;10:419–24.
20. Martinez-Rumayor A, Richards AM, Burnett JC, Januzzi JL Jr. Biology
of the Natriuretic Peptides. Am J Cardiol. 2008 Feb;101(3):S3–S8.
21. Kate ten CA, Tibboel D, Kraemer US. B-type natriuretic peptide as a
parameter for pulmonary hypertension in children. A systematic review.
Eur J Pediatr. 2015 Aug 23;174(10):1267–75.
22. Takatsuki S, Wagner BD, Ivy DD. B-type natriuretic peptide and
amino-terminal pro-B-type natriuretic peptide in pediatric patients with
pulmonary arterial hypertension. Congenit Heart Dis. Blackwell
Publishing Inc; 2012 May;7(3):259–67.
23. Hauser JA, Demyanets S, Rusai K, Goritschan C, Weber M, Panesar D,
et al. Diagnostic performance and reference values of novel biomarkers
of paediatric heart failure. Heart. 3rd ed. 2016 Sep 27;102(20):1633–9.
24. Mladosievicova B, Urbanova D, Radvanska E, Slavkovsky P, Simkova I.
Role of NT-proBNP in detection of myocardial damage in childhood
leukemia survivors treated with and without anthracyclines. Journal of
Experimental & Clinical Cancer Research; 2012 Oct 11;31(1):1–1.
25. Li Q, Cui C-Y, Zhang C, Guo S-Y, Zhang Q. On the changes of
NT-proBNP level in children having undergone radical operation of
tetralogy of Fallot and the clinical significance. Eur Rev Med Pharmacol
Sci. 2015 Aug;19(16):3018–22.
26. Balaguer M, Alejandre C, Vila D, Esteban E, Carrasco JL, Cambra FJ,
et al. Bronchiolitis Score of Sant Joan de Déu: BROSJOD Score,
validation and usefulness. Pediatr Pulmonol. 2nd ed. 2017
Apr;52(4):533–9.
27. Lai WW, Geva T, Shirali GS, Frommelt PC, Humes RA, Brook MM, et al.
Guidelines and Standards for Performance of a Pediatric Echocardiogram:
A Report from the Task Force of the Pediatric Council of the American
Society of Echocardiography. Journal of the American Society of
Echocardiography. 2006 Dec;19(12):1413–30.
28. Koestenberger M, Ravekes W, Everett AD, Stueger HP, Heinzl B,
Gamillscheg A, et al. Right ventricular function in infants, children
and adolescents: reference values of the tricuspid annular plane
systolic excursion (TAPSE) in 640 healthy patients and calculation of z
score values. J Am Soc Echocardiogr. 2009 Jun;22(6):715–9.
29. Tei C, Dujardin KS, Hodge DO, Bailey KR, McGoon MD, Tajik AJ, et al.
Doppler echocardiographic index for assessment of global right
ventricular function. Journal of the American Society of
Echocardiography. 1996 Nov;9(6):838–47.
30. Koestenberger M, Apitz C, Abdul-Khaliq H, Hansmann G. Transthoracic
echocardiography for the evaluation of children and adolescents with
suspected or confirmed pulmonary hypertension. Expert consensus
statement on the diagnosis and treatment of paediatric pulmonary
hypertension. The European Paediatric Pulmonary Vascular Disease
Network, endorsed by ISHLT and D6PK. Heart. BMJ Publishing Group Ltd and
British Cardiovascular Society; 2016 May;102 Suppl 2(Suppl 2):ii14–22.
31. Habash S, Laser KT, Moosmann J, Reif R, Adler W, Glöckler M, et al.
Normal values of the pulmonary artery acceleration time (PAAT) and the
right ventricular ejection time (RVET) in children and adolescents and
the impact of the PAAT/RVET-index in the assessment of pulmonary
hypertension. Int J Cardiovasc Imaging. Springer Netherlands; 2019
Feb;35(2):295–306.
32. Abraham S, Weismann CG. Left Ventricular End-Systolic Eccentricity
Index for Assessment of Pulmonary Hypertension in Infants.
Echocardiography. 2016 Jan 16;33(6):910–5.
33. Cui W, Roberson DA. Left ventricular Tei index in children:
comparison of tissue Doppler imaging, pulsed wave Doppler, and M-mode
echocardiography normal values. J Am Soc Echocardiogr. 2006
Dec;19(12):1438–45.
34. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under
two or more correlated receiver operating characteristic curves: a
nonparametric approach. Biometrics. 1988 Sep;44(3):837–45.
35. Koo TK, Li MY. A Guideline of Selecting and Reporting Intraclass
Correlation Coefficients for Reliability Research. J Chiropr Med. 2016
Jun;15(2):155–63.
36. Cui W, Roberson DA, Chen Z, Madronero LF, Cuneo BF. Systolic and
diastolic time intervals measured from Doppler tissue imaging: normal
values and Z-score tables, and effects of age, heart rate, and body
surface area. J Am Soc Echocardiogr. 2008 Apr;21(4):361–70.
37. Burkett DA, Slorach C, Patel SS, Redington AN, Ivy DD, Mertens L, et
al. Left Ventricular Myocardial Function in Children With Pulmonary
Hypertension: Relation to Right Ventricular Performance and
Hemodynamics. Circ Cardiovasc Imaging. American Heart Association, Inc;
2015 Aug;8(8):e003260–19.
38. Haeck MLA, Höke U, Marsan NA, Holman ER, Wolterbeek R, Bax JJ, et
al. Impact of right ventricular dyssynchrony on left ventricular
performance in patients with pulmonary hypertension. Int J Cardiovasc
Imaging. Springer Netherlands; 2014 Feb 4;30(4):713–20.
39. Bekhof J, Reimink R, Brand PLP. Systematic review: Insufficient
validation of clinical scores for the assessment of acute dyspnoea in
wheezing children. Paediatric Respiratory Reviews. 2014
Mar;15(1):98–112.
40. Destino L, Weisgerber MC, Soung P, Bakalarski D, Yan K, Rehborg R,
et al. Validity of respiratory scores in bronchiolitis. Hosp Pediatr.
2012 Oct;2(4):202–9.
41. Brown PM, Schneeberger DL, Piedimonte G. Biomarkers of respiratory
syncytial virus (RSV) infection: specific neutrophil and cytokine levels
provide increased accuracy in predicting disease severity. Paediatric
Respiratory Reviews. 2015 Sep;16(4):232–40.
42. Takeuchi D, Saji T, Takatsuki S, Fujiwara M. Abnormal tissue doppler
images are associated with elevated plasma brain natriuretic peptide and
increased oxidative stress in acute Kawasaki disease. Circ J. 2007
Mar;71(3):357–62.
43. Ozde C, Dogru M, Ozde Ş, Kayapinar O, Kaya A, Korkmaz A. Subclinical
right ventricular dysfunction in intermittent and persistent mildly
asthmatic children on tissue Doppler echocardiography and serum
NT-proBNP: Observational study. Pediatr Int. John Wiley & Sons, Ltd
(10.1111); 2018 Nov;60(11):1024–32.
44. Mikkelsen KV, Møller JE, Bie P, Ryde H, Videbaek L, Haghfelt T. Tei
index and neurohormonal activation in patients with incident heart
failure: serial changes and prognostic value. European Journal of Heart
Failure. John Wiley & Sons, Ltd; 2006 Oct;8(6):599–608.