Review articles
Herz
https://doi.org/10.1007/s00059-019-4825-4
Received: 14 December 2018
Revised: 5 April 2019
Accepted: 21 May 2019
© Springer Medizin Verlag GmbH, ein Teil von
Springer Nature 2019
Electronic supplementary
material
The online version of this article (https://doi.
org/10.1007/s00059-019-4825-4) contains
supplementary material, which is available to
authorized users.
Cardiomyopathies, or heart muscle dis-
orders, include four primary types:
dilated cardiomyopathy (DCM), hy-
pertrophic cardiomyopathy (HCM),
restrictive cardiomyopathy (RCM), and
arrhythmogenic right ventricular car-
diomyopathy (ARVC) according to the
classification of World Health Orga-
nization (WHO; [1, 2]). Dilated car-
diomyopathy is a heart muscle disorder
characterized by dilation and systolic
impairment of the left ventricle or both
ventricles in the absence of hyperten-
sion, coronary artery disease, or valvular
abnormalities [3–5]. It usually results in
a large cost burden because of the high
rate of hospitalization and the potential
need for heart transplantation [4]. Di-
lated cardiomyopathy is one of the main
risk factors of sudden cardiac death and
heart failure [4, 5]. Apart from acquired
factors (such as infections, toxins, nu-
tritional deficiencies, or autoimmune
disorders), idiopathic factors are equally
important in the pathogenesis leading to
DCM [3, 5]. A large number of studies
reveal that idiopathic DCM, including
familial and sporadic cases, has a genetic
cause [2, 5–9].
H.-J. Fang1 · B.-P. Liu2
1
Department of Pediatrics, Shandong Provincial Hospital Affiliated to Shandong University, Jinan, China
2
Department of Epidemiology, Shandong University School of Public Health, Jinan, China
Prevalence of TTN mutations
in patients with dilated
cardiomyopathy
A meta-analysis
TTN encoding the protein of titin
consists of 363 exons [6, 10]. Titin is
the largest protein (34,350 amino acids;
[11]); it provided an external scaffold and
interacts with both thin and thick fila-
ments, and plays an important role in
sarcomere assembly and in force gener-
ation [5]. N2B or N2BA are isoforms of
titin in the heart [12, 13]. It has been
proved that TTN mutations, especially in
the A-band region, are associated with
DCM [10, 11, 14, 15]. Herman et al.
found that the prevalence of TTN muta-
tions is significantly higher among DCM
patients than among the general popu-
lation [12]. Besides, some studies reveal
that the prevalence of FDCM is higher
than SDCM [12, 16].
Previous studies have focused on
the prevalence of TTN mutations in
DCM patients. However, the prevalence
varies depending on the populations. To
our knowledge, no meta-analysis of the
prevalence of TTN mutations in DCM
patients is available. In this study, our
aims were to estimate the prevalence of
TTN mutations in DCM patients. In
addition, FDCM and SDCM patients
were also assessed in this study.
Methods
This study was conducted according
to PRISMA guidelines (http://www.
prisma-statement.org/). We collected
published research by searching several
databases (PubMed, ISI Web of Science,
EMBASE, China National Knowledge In-
frastructure) using the following search
items: “TTN” or “TTN mutations” or
“titin” and “dilated cardiomyopathy” or
“DCM”. Two investigators (HJF and
BPL) independently identified relevant
reports published from the date of set-
ting up the databases to 30 August, 2018.
Details of the search process and study
selection are shown in . Fig. 1. Besides,
we reviewed the references of the in-
cluded studies and systematic reviews to
identify additional studies that were not
captured by our database searches.
Aninitial screening was performed ac-
cording to the following criteria: (1) no
overlap in subjects or data from the same
study population; (2) only human sub-
jects; (3) numbers of DCM patients with
and without TTN mutations, or data to
help calculate the numbers, should be
reported.
Information extracted from the stud-
ies included the title, name of the first
author, year of publication, country of
study population, qualitative description
of target population, and sequencing
methods. More details are provided in
. Table 1. For the meta-analyses, we
specifically extracted the total number
of DCM patients, number of DCM pa-
tients with TTN mutations, number of
FDCM patients with TTN mutations,
and number of SDCM patients with TTN
mutations. The quality of the prevalence
study was evaluated according to the
Agency for Healthcare Research and
Quality (AHRQ; [17]). All studies were
assessed individually according to each
item and the overall scores were obtained
(five items with reverse scoring).
Herz
 Table 1 Demographic characteristics of study populations and criteria used to define DCM

Herz
Study Country Year Male/female Mean age Identification Criteria of DCM and FDC
(years) LVEDD LVEF FDC
Akinrinade (2015) [10] Finland 2015 107/38 45.7 TTN truncating variants >27 mm/m2 <45% Family history of hypertrophic cardiomyopathy
(HCM), DCM, or arrhythmogenic right ventricular
cardiomyopathy (ARVC), or two or more patients
Review articles

with atrial fibrillation before age 40 or rhythm/

conduction disturbances requiring pacemakers in
the family
Dal Ferro (2017) [22] Italy 2017 111/41 41.0 TTN truncating variants NA <50% All familial cases fulfilled the published criteria
Fatkin (2016) [14] America 2016 54/28 NA TTN truncating variants NA NA NA
Felkin (2016) [25] UK 2016 58/12 33.5 TTN truncating variants NA NA Family history
Franaszczyk (2017) Poland 2017 48/24 34.0 TTN truncating variants >117% of <45% Two subjects in the family met the same diagnostic
[20] the predicted value criteria for DCM as proband
Haas (2015) [16] Germany 2015 405/212 NA TTN truncating variants >117% of the pre- <45% Two subjects in the family met the same diagnostic
dicted value criteria for DCM as proband
Herman (2012) [12] America 2012 208/104 39.0 TTN truncating variants NA NA NA
Janin (2017) [28] France 2017 NA NA TTN truncating variants NA NA At least 2 first-degree relatives in the same family
were affected
Klauke (2017) [23] Germany 2017 22/8 NA Variants classification NA NA Family history
Pugh (2014) [29] America 2014 404/330 NA Variants classification NA NA NA
Roberts (2015) [26] UK 2015 NA NA TTN truncating variants LVDD >56 mm <50% A positive family history (DCM or sudden explained
cardiac death in ≥2 family members) and three had
subclinical skeletal myopathy (elevated creatine
kinase blood levels)
Tayal (2017) [24] UK 2017 469/247 53.5 TTN truncating variants NA NA Family history
Tobita (2018) [27] Japan 2018 99/21 39.1 TTN truncating variants More than the av- <50% At least 1 additional affected family member with
erage value of the any cardiomyopathy or patients with family history
healthy Japanese of sudden cardiac death
population
DCM dilated cardiomyopathy, FDCM familial dilated cardiomyopathy, LVEDD left ventricular end-diastolic dimension, LVEF left ventricular ejection fraction, NA not available

All analyses were performed in R ver-
sion 3.4.3 (R Foundation for Statistical
Computing, Vienna, Austria) and sta-
tistical significance was indicated at the
level of p = 0.05. We computed all pooled
prevalence rates and 95% confidence in-
tervals (CIs) using logit transformation.
Inverse variance weighting was used to
calculate pooled estimates. Homogene-
ity in the prevalence was tested with
Q and I2 statistics. If the heterogeneity
among the studies was large (I2 > 50%),
random-effect models (REM) were used
to calculate the pooled estimates, oth-
erwise (I2 < 50%), fixed-effect models
(FEM) were applied. The prevalence
of TTN mutations in DCM, FDCM,
and SDCM patients was analyzed sep-
arately. Meta-regression analyses were
performed to assess the factors leading
to the homogeneity. Influence analyses
were also conducted to validate the sta-
bility of the data by omitting each study.
A study was suspected to excessively
influence the final point estimation if
its omission led to an estimate beyond
the 95% CI of the combined analysis.
Cumulative meta-analyses associated
with total sample size and publication
time (year) were calculated to find the
changing trends in prevalence. Publi-
cation bias was evaluated with Egger’s
regression asymmetry test [18], Begg’s
adjusted correlation rank test [19], and
funnel plots.
Results
An initial search yielded 1524 studies
from databases and three studies from
other sources. After removing duplicates,
1219 studies were retained for analysis.
The PRISMA flow diagram for the se-
lection process is shown in . Fig. 1. We
found 13 potential articles after removing
studies according to our aforementioned
criteria. In total, the analysis included
4018 DCM patients, 702 FDCM patients,
and 1571 SDCM patients. The demo-
graphic characteristics and the echocar-
diographic criteria of defining DCM and
FDCM for each study are presented in
. Table 1. These data were not available
for all studies. Using the AHRQ method
to evaluate the quality of these studies,
one study [20] had eight stars, four stud-
Abstract · Zusammenfassung
Herz https://doi.org/10.1007/s00059-019-4825-4
© Springer Medizin Verlag GmbH, ein Teil von Springer Nature 2019
H.-J. Fang · B.-P. Liu
Prevalence of TTN mutations in patients with dilated
cardiomyopathy. A meta-analysis
Abstract
A meta-analysis was performed to assess the
prevalence of TTN mutations in patients with
dilated cardiomyopathy (DCM). Prevalence
point estimates and 95% confidence
intervals were computed using the logit
transformation formula. The prevalence
of TTN mutations in patient with DCM,
familial dilated cardiomyopathy (FDCM), and
sporadic dilated cardiomyopathy (SDCM)
was 0.17 (95% CI: 0.14–0.19), 0.23 (95% CI:
0.20–0.26), and 0.16 (95% CI: 0.12–0.21),
respectively. No individual study had a
marked influence on the pooled prevalence
in the meta-analysis. Meta-regression analysis
Prävalenz von TTN-Mutationen bei Patienten mit dilatativer
Kardiomyopathie. Eine Metaanalyse
Zusammenfassung
Eine Metaanalyse wurde durchgeführt,
um die Prävalenz von TTN-Mutationen bei
Patienten mit dilatativer Kardiomyopathie
(DCM) zu erfassen. Prävalenz-Punktschätzung
und 95 %-Konfidenzintervalle wurden unter
Verwendung der Logit-Transformationsformel
errechnet. Die Prävalenz von TTN-Mutationen
bei DCM-Patienten betrug 0,17 (95 %-
KI: 0,14–0,19), bei familiärer dilatativer
Kardiomyopathie (FDCM) 0,23 (95 %-KI:
0,20–0,26) bzw. bei sporadischer dilatativer
Kardiomyopathie (SDCM) 0,16 (95 %-KI:
0,12–0,21). Es gab keine einzelne Studie
mit deutlichem Einfluss auf die gepoolte
Prävalenz im Rahmen der Metaanalyse.
Die Metaregressionsanalyse zwischen dem
ies [21–24] had seven stars, four studies
[16, 25–27] had six stars, and four studies
[12, 14, 28, 29] had five stars. More de-
tails can be found in the supplementary
online Table 1.
We analyzed all 13 studies with
an REM, which revealed a pooled preva-
lence of 0.17 (95% CI: 0.14–0.19) and an
I2 value of 73% for the DCM patients.
In addition, eight studies were analyzed
with an FEM revealing a pooled preva-
lence of 0.23 (95% CI: 0.20–0.26) and an
I2 value of 26% for the FDCM patients.
Eight studies analyzed with an REM
revealed a pooled prevalence of 0.16
between the logit event for prevalence and
sample size explained 32% of between-study
variance (p < 0.05). Cumulative meta-analysis
confirmed the influence of sample size on
the reported prevalence among the different
studies. In conclusion, the present analysis
suggests that TTN mutations are familial in
DCM patients. More attention should be paid
to TTN mutations in clinical examinations.
Keywords
Genes · Titin · Cardiac muscle · Heart failure ·
Systematic review
Logit-Ereignis für die Prävalenz und der
Stichprobengröße erklärte 32 % der Varianz
zwischen den Studien (p < 0,05). Die kumu-
lative Metaanalyse bestätigte den Einfluss
der Stichprobengröße auf die angegebene
Prävalenz bei den verschiedenen Studien. Die
Schlussfolgerung der vorliegenden Analyse
war, dass TTN-Mutationen familiär gehäuft
bei DCM-Patienten auftreten. Daher sollte
mehr Aufmerksamkeit auf TTN-Mutationen
bei der klinischen Untersuchung gelegt
werden.
Schlüsselwörter
Gene · Titin · Herzmuskel · Herzinsuffizienz ·
Systematische Übersicht
(95% CI: 0.12–0.21) and an I2 value of
73% for the SDCM patients. Detailed
results of these meta-analyses are shown
in . Fig. 2.
In the influence analyses, no individ-
ual study markedly influenced the pooled
prevalence in the meta-analysis of DCM
patients. Detailed results of these meta-
analyses are shown in online Fig. 1. Meta-
regression analyses between logit preva-
lence and the number of DCM patients
explained 31.55% of between-study vari-
ance (p = 0.026), with a logit prevalence
decrease of 0.0007 for one DCM pa-
tient. The meta-regression results for the
Herz
 Review articles

tions. A meta-analysis reported that the

frequency of the LMNA, PLN, RBM20,
Idenficaon

Records idenfied through Addional records idenfied MYBPC3, MYH7, TNNT2, and TNNI3
database searching through other sources mutations was 5%, 2%, 2%, 4%, 3%, 2%,
(n =1524) (n =3 )
and 1%, respectively [8]. Besides, TTN
mutations had a higher prevalence in pa-
tients with DCM compared with other
Records aer duplicates removed types of cardiomyopathy and compared
(n =1219)
with the general population [10, 12, 20].
Screening

Many researchers used pedigree stud-

ies to prove the positive relationship
Records screened Records excluded between TTN mutations and FDCM
(n =1219) (n =1169) [30–32]. Our results were unsurprising
since many studies and reviews have
focused on FDCM [2, 9, 15]. However,
Full-text arcles assessed Full-text arcles excluded, in this study, we found that the preva-
Eligibility

for eligibility with reasons

(n =50) (n =37)
lence of TTN mutations in patients with
-review arcles (n=16) SDCM was relatively high (16%). Thus,
-not related topics or we should pay attention to SDCM since
insufficient informaon
Studies included in (n=20) it can be concealed.
qualitave synthesis -reported same research The heterogeneity of the meta-analy-
(n =13) data (n=1)
ses associated with DCM and SDCM was
relatively high in this study (I2 = 73%).
Included

Althoughinfluence analyses revealed that

Studies included in
quantave synthesis no individual study markedly influenced
(meta-analysis) the pooled prevalence, there were two
(n =13)
studies [14, 23] that reported a relatively
high/low prevalence leading to the high
heterogeneity. Klauke et al. included
Fig. 1 8 PRISMA flow diagram showing process of study selection for inclusion in our meta-analyses.
(From Moher et al. [33]). For more information, visit www.prisma-statement.org 30 DCM patients with non-ischemic se-
vere heart failure needing heart trans-
plantation (HTx) or mechanical circula-
number of SDCM patients were not sig- we also found the reported prevalence tory support and transferred to the center
nificant. In addition, the meta-regres- had a decreasing trend when year of pub- for HTx evaluation [23]. These patients
sion results for year of publication were lication was considered. However, this with advanced or end-stage cardiomy-
not significant for DCM/FDCM patients. trend could not be verified by meta-re- opathy might be different from DCM pa-
The sensitivity analysis revealed high het- gression. The relationship between year tients included in other studies. Besides,
erogeneity between different TTN mu- of publication and prevalence of TTN we could not find a time period used for
tations associated with TTN truncating mutations is shown in . Fig. 4b. patient identification. These differences
variants (I2 = 60%) and variant classifica- might cause the high prevalence (47%).
tion (I2 = 95%; online Fig. 2). The funnel Discussion In addition, there were two studies using
plots were symmetrical for three meta- variant classification (including variant of
analyses (. Fig. 3). No publication bias In this meta-analysis, the overall preva- uncertain significance, likely pathogenic
was found in the studies of DCM patients lence of TTN mutations in patients with and pathogenic) for the identification of
using Egger’s test (t = 1.58, p = 0.141) and DCM was 17%. In addition, we found TTN mutations. Other studies reported
Begg’s test (z = 1.34, p = 0.180). There was a higher prevalence of the mutation in TTN truncating variants. However, the
also no publication bias in the studies of patients with FDCM than in those with two methods of identification of TTN
FDCM and SDCM patients according to SDCM (23% vs. 16%, respectively). In mutations both had high heterogeneity.
analyses with Egger’s test and Begg’s test. this study, we also found that the sample Fatkin et al. included DCM patients aged
Cumulative meta-analyses confirmed the size of the studies was negatively corre- 21 years or younger at the time of the
influence of the sample size of the studies lated with the prevalence of TTN muta- DCM diagnosis [14]. Different popula-
on the reported prevalence of TTN mu- tions. tions might contribute to the high hetero-
tations among different studies. As the TTN mutations have an important ef- geneity. In this study, the meta-analysis of
sample size increased, the prevalence de- fect on DCM [5]. The frequency of other FDCM patients showed a relatively small
creased and was gradually stable. More familiar genes proved to be associated heterogeneity (I2 = 26%). In contrast to
details are shown in . Fig. 4a. In addition, with DCM is lower than TTN muta- SDCM, normal diagnostic criteria have

Herz
Fig. 2 9 Forest plots for
prevalence of TTN mu-
tations in the 13 studies:
a dilated cardiomyopa-
thy patients; b familial
dilated cardiomyopathy
patients; c sporadic dilated
cardiomyopathy patients

Herz
 Review articles

been applied to FDCM in recent years

[9]. Considering the existing inaccurate
variant assessment strategies, the clinical
interpretation of TTN truncating vari-
ants might be affected across the general
population. This might be another rea-
son for the high heterogeneity among the
population of SDCM patients.
The sample size was found to be asso-
ciated with the prevalence of TTN muta-
tions in this study. Meta-regression anal-
ysis also showed that larger sample sizes
were associated with a lower prevalence
of TTN mutations in this study. Large
sample sizes might yield stable results. In
small samples, information bias can have
a significant effect on the results. At the
same time, we also found a decreasing
trend in the prevalence associated with
year of publication. This might be due to
the more correct use of diagnostic criteria
for TTN mutations and DCM.

Limitations
This meta-analysis has some limitations.
First, different types (e.g., frameshift,
nonsense, essential splice site variants)
and areas (e.g., A-band, I-band, Z-band,
and M-band) of TTN mutations were
not analyzed in this study. Although the
types and areas are important to TTN
mutations, our main aim in this study
was to explore the total prevalence of
TTN mutations in DCM patients. Sec-
ond, interpretation and extrapolation of
the conclusions should be made with
care considering the large number of
included studies associated with TTN
truncated variants. Detailed analyses of
TTN mutations are needed in clinical
practice and in research. Third, the
number of articles was relatively small,
which might have limited the gener-
Fig. 3 8 Funnel plots of 13 studies included in the meta-analysis of the prevalence of TTN mutations: alizability of our conclusions. Fourth,
a dilated cardiomyopathy patients; b familial dilated cardiomyopathy patients; c sporadic dilated car- the studies included in this analysis did
diomyopathy patients
not specifically investigate the effect of
race, gender, and age because of limited
data. However, these factors are impor-
tant for genetic studies. Finally, only
publications in English were included
in the meta-analysis, and we could not
exclude the possibility of publication
bias influencing the results of this meta-
analysis, even though statistical analysis
indicated no publication bias.

Herz

Fig. 4 8 Cumulative meta-analyses of relationships between sample size (a), publication year (b), and
the prevalence of TTN mutations in patients with dilated cardiomyopathy. These studies were added
one at a time in specific order
Conclusion
In conclusion, this meta-analysis sug-
gests that TTN mutations in patients
with dilated cardiomyopathy are fa-
milial. There is a higher prevalence of
TTN mutations in patients with familial
dilated cardiomyopathy than in those
with sporadic dilated cardiomyopathy.
Since potential biases and confounders
could not be ruled out completely in
this meta-analysis, further studies are
needed to confirm these results.
Corresponding address
B.-P. Liu, MPH
Department of Epidemiology, Shandong
University School of Public Health
No.44, Wenhuaxi Rd, 250012 Jinan, China
sduliubaopeng@163.com
Author Contribution. Bao-Peng Liu and Hui-Juan
Fang conceived of the study. Bao-Peng Liu and Hui-
Juan participated in the data analysis. Bao-Peng Liu
and Hui-Juan wrote, reviewed, and approved the
final manuscript.
Compliance with ethical
guidelines
Conflict of interest H.-J. Fang and B.-P. Liu declare
that they have no competing interests.
For this article no studies with human participants
or animals were performed by any of the authors. All
studies performed were in accordance with the ethical
standards indicated in each case.
References
1. Richardson P, McKenna W, Bristow M et al (1996)
Report of the 1995 World Health Organization/
International Society and Federation of Cardiology
Task Force on the Definition and Classification of
cardiomyopathies. Circulation 93(5):841–842
2. Menon SC, Olson TM, Michels VV (2008) Genetics
of familial dilated cardiomyopathy. Prog Pediatr
Cardiol 25(1):57–67. https://doi.org/10.1016/j.
ppedcard.2007.11.013
3. Elliott P, Andersson B, Arbustini E et al (2008)
Classification of the cardiomyopathies: a position
statement from the European Society Of Cardiol-
ogy Working Group on Myocardial and Pericardial
Diseases. Eur Heart J 29(2):270–276. https://doi.
org/10.1093/eurheartj/ehm342
4. Jefferies JL, Towbin JA (2010) Dilated cardiomy-
opathy. Lancet 375(9716):752–762. https://doi.
org/10.1016/S0140-6736(09)62023-7
5. Garcia-Pavia P, Cobo-Marcos M, Guzzo-Merello G
et al (2013) Genetics in dilated cardiomyopathy.
Biomark Med 7(4):517–533. https://doi.org/10.
2217/bmm.13.77
6. Cho KW, Lee J, Kim Y (2016) Genetic variations
leading to familial dilated cardiomyopathy. Mol
Cells 39(10):722–727. https://doi.org/10.14348/
molcells.2016.0061
7. Garfinkel AC, Seidman JG, Seidman CE (2018)
Genetic pathogenesis of hypertrophic and dilated
cardiomyopathy. Heart Fail Clin 14(2):139–146.
https://doi.org/10.1016/j.hfc.2017.12.004
8. Kayvanpour E, Sedaghat-Hamedani F, Amr A
et al (2017) Genotype-phenotype associations
in dilated cardiomyopathy: meta-analysis on
more than 8000 individuals. Clin Res Cardiol
106(2):127–139. https://doi.org/10.1007/s00392-
016-1033-6
9. Petretta M, Pirozzi F, Sasso L, Paglia A, Bonaduce D
(2011) Review and metaanalysis of the frequency
of familial dilated cardiomyopathy. Am J Cardiol
108(8):1171–1176. https://doi.org/10.1016/j.
amjcard.2011.06.022
10. Akinrinade O, Koskenvuo JW, Alastalo T-P (2015)
Prevalence of Titin truncating variants in general
population. PLoS ONE. https://doi.org/10.1371/
journal.pone.0145284
11. Fatkin D, Huttner IG (2017) Titin-truncating
mutations in dilated cardiomyopathy: the long
and short of it. Curr Opin Cardiol. https://doi.org/
10.1097/hco.0000000000000382
12. Herman DS, Lam L, Taylor MR et al (2012) Trun-
cations of titin causing dilated cardiomyopathy.
N Engl J Med 366(7):619–628. https://doi.org/10.
1056/NEJMoa1110186
13. Guo W, Bharmal SJ, Esbona K, Greaser ML (2010)
Titin diversity-alternative splicing gone wild.
J Biomed Biotechnol. https://doi.org/10.1155/
2010/753675
14. Fatkin D, Lam L, Herman DS et al (2016) Titin
truncating mutations: a rare cause of dilated
cardiomyopathy in the young. Prog Pediatr Cardiol
40:41–45. https://doi.org/10.1016/j.ppedcard.
2016.01.003
15. Hershberger RE, Siegfried JD (2011) Update
2011: clinical and genetic issues in familial
dilated cardiomyopathy. J Am Coll Cardiol
57(16):1641–1649. https://doi.org/10.1016/j.jacc.
2011.01.015
16. HaasJ,FreseKS,PeilBetal(2015)Atlasoftheclinical
genetics of human dilated cardiomyopathy. Eur
Heart J 36(18):1123–U1143. https://doi.org/10.
1093/eurheartj/ehu301
17. Alaa Rostom, Catherine Dubé, Ann Cranney et
al. Celiac Disease. Rockville (MD): Agency
for Healthcare Research and Quality (US); 2004
Sep. (Evidence Reports/Technology Assessments,
No. 104.) Appendix D. Quality Assessment
Herz
 Review articles

Forms. https://www.ncbi.nlm.nih.gov/books/ for systematic reviews and meta-analyses: the

NBK35149/. PRISMA statement. Plos Med 6(6):e1000097.
18. Egger M, Davey Smith G, Schneider M, Minder C https://doi.org/10.1371/journal.pmed1000097
(1997) Bias in meta-analysis detected by a simple,
graphical test. BMJ 315(7109):629–634
19. BeggCB, MazumdarM(1994)Operatingcharacter-
istics of a rank correlation test for publication bias.
Biometrics 50(4):1088–1101
20. Franaszczyk M, Chmielewski P, Truszkowska G
et al (2017) Titin truncating variants in dilated
cardiomyopathy—prevalence and genotype-
phenotype correlations. PLoS ONE 12(1):e169007.
https://doi.org/10.1371/journal.pone.0169007
21. Akinrinade O, Ollila L, Vattulainen S et al (2015)
Genetics and genotype-phenotype correlations
in Finnish patients with dilated cardiomyopathy.
Eur Heart J 36(34):2327–2337. https://doi.org/10.
1093/eurheartj/ehv253
22. Dal Ferro M, Stolfo D, Altinier A et al (2017) Associa-
tion between mutation status and left ventricular
reverse remodelling in dilated cardiomyopathy.
Heart 103(21):1704–1710. https://doi.org/10.
1136/heartjnl-2016-311017
23. Klauke B, Gaertner-Rommel A, Schulz U et al
(2017) High proportion of genetic cases in patients
with advanced cardiomyopathy including a novel
homozygous Plakophilin 2-gene mutation. PLoS
ONE 12(12):e189489. https://doi.org/10.1371/
journal.pone.0189489
24. Tayal U, Newsome S, Buchan R et al (2017)
Phenotype and clinical outcomes of Titin car-
diomyopathy. J Am Coll Cardiol 70(18):2264–2274.
https://doi.org/10.1016/j.jacc.2017.08.063
25. Felkin LE, Walsh R, Ware JS et al (2016) Recovery of
cardiacfunctionincardiomyopathycausedbyTitin
truncation. JAMA Cardiol 1(2):234–235. https://
doi.org/10.1001/jamacardio.2016.0208
26. Roberts AM, Ware JS, Herman DS et al (2015)
Integrated allelic, transcriptional, and phenomic
dissection of the cardiac effects of titin truncations
inhealthanddisease. SciTranslMed7(270):270ra6.
https://doi.org/10.1126/scitranslmed.3010134
27. Tobita T, Nomura S, Fujita T et al (2018) Genetic
basis of cardiomyopathy and the genotypes
involved in prognosis and left ventricular reverse
remodeling. Sci Rep 8(1):1998. https://doi.org/10.
1038/s41598-018-20114-9
28. Janin A, N’Guyen K, Habib G et al (2017) Truncating
mutations on myofibrillar myopathies causing
genes as prevalent molecular explanations on
patients with dilated cardiomyopathy. Clin
Genet 92(6):616–623. https://doi.org/10.1111/
cge.13043
29. Pugh TJ, Kelly MA, Gowrisankar S et al (2014)
The landscape of genetic variation in dilated
cardiomyopathy as surveyed by clinical DNA
sequencing. Genet Med 16(8):601–608. https://
doi.org/10.1038/gim.2013.204
30. Siu BL, Niimura H, Osborne JA et al (1999)
Familial dilated cardiomyopathy locus maps to
chromosome 2q31. Circulation 99(8):1022–1026
31. Gerull B, Atherton J, Geupel A et al (2006)
Identification of a novel frameshift mutation in the
giant muscle filament titin in a large Australian
family with dilated cardiomyopathy. J Mol Med
84(6):478–483. https://doi.org/10.1007/s00109-
006-0060-6
32. Gerull B, Gramlich M, Atherton J et al (2002) Muta-
tions of TTN, encoding the giant muscle filament
titin, cause familial dilated cardiomyopathy. Nat
Genet 30(2):201–204. https://doi.org/10.1038/
ng815
33. Moher D, Liberati A, Tetzlaff J, Altman DG, The
PRISMA Group (2009) Preferred reporting items

Herz