† These authors contributed equally.
Background: Duchenne muscular dystrophy (DMD) is major childhood
muscular dystrophy. Pre-implantation genetic testing (PGT) is an alternative to
prenatal diagnosis. This study performed SNP microarray with karyomapping PGT of
DMD in comparison to PCR-based techniques for validation. Methods: Two
families at risk of having DMD offspring decided to have karyomapping PGT. PCR
protocol using mini-sequencing and intragenic microsatellites-based linkage
analysis was developed and applied. Results: Karyotyping results of
family DA (DMD c.895G
Duchenne muscular dystrophy (DMD, MIM #310200, Xp21.2-p21.1), an X-linked recessive disorder, is one of the most common muscular dystrophies in childhood with the prevalence of 2.9 per 10,000 live male births [1]. DMD gene possesses 79 exons covering over 2.3 Mb. DMD is associated with loss or abnormal function of the key muscular protein called dystrophin, one of spectrin protein superfamily with 427 kilodalton (kDa), and it is expressed in skeletal muscles, cardiac muscles, cerebral cortical neurons and Purkinje cerebellar neurons [2, 3]. Mutations within DMD gene cause deficiency of dystrophin and are associated with muscular membrane instability [4], leading to apoptosis and necrosis of muscle cells. DMD patients usually express gross motor delay, proximal muscle weakness, calf hypertrophy (at 3 to 5 years old), and loss of ambulation (by the age of 12 to 13) [5]. The patients suffer progressive proximal muscle weakness and are at increased risk of death from respiratory failure and cardiomyopathy occurring in their 30’s [6, 7]. Gross deletion (60–65%) and duplication (8–15%) which lead to frameshift are major mutations of the DMD gene while the remainder are caused from point mutations [8, 9, 10] with approximately one-third of DMD mutations being de novo [3].
Mutations of DMD gene cause dystrophin protein deficiency and absence of
dystrophin-associated glycoprotein complex (DGC). Following the dysregulated
signaling pathways and abnormal membrane structure, the reactive oxygen species
(ROS) develop and lead mitochondrial dysfunction. The increased intracellular
Ca
Similar to other chronic and progressive conditions, the patients and parents encounter psychologic and social difficulties. Affected children suffer from diseases associated with discomfort and pain, restricted physical activities, school absences and learning delay. The parents worry about disease advancement and disability. The parent may need to leave their job in order to take care of affected child leading to a decrease in quality of life and income loss. Additional expenses such as nursing, food, school support, and transport are greater than for a healthy child [1]. The patients suffer progressive proximal muscle weakness and die from respiratory failure and cardiomyopathy in their fourth decade of life [6, 7].
Since there is no specific and curative treatment available, genetic counselling and prenatal diagnosis are recommended. Female carrier identification can be done using a commercial DMD mutation screening test by multiplex ligation-dependent probe amplification (MLPA) [12]. A primary test is carried out to detect large deletions or duplication within the DMD gene. If the primary test is negative, sequence analysis of DMD encoding region is performed to look for minor frameshift or nonsense mutations using denaturing gradient gel electrophoresis (DGGE), protein truncation test and Sanger sequencing [13, 14]. For those without an answer, next generation sequencing (NGS) can be used to check for all possible mutations [15].
Early prenatal diagnosis (PND) of DMD employed fetal blood sampling and the measurement of plasma creatine phosphokinase activity [16]. However, this technique is not specific [17]. First trimester fetal sex deter mination was employed in pregnancies at risk for DMD [18]. In the early molecular era, prenatal diagnosis and carrier detection of DMD utilized closely linked restriction fragment length polymorphism (RFLPs) markers [19]. Linkage analysis based using dinucleotide repeat polymorphisms was also applied for PND of DMD [20]. In utero fetal muscle biopsy for PND of DMD was performed in cases of problems with molecular genetic analysis [21, 22].
First trimester chorionic villous sampling (CVS) [23], second trimester amniocentesis [24] and fetal blood sampling (FBS or cordocentesis) [25] are choices of invasive PND procedures for prenatal diagnosis. Use of non-invasive cell free fetal DNA (cffDNA) for monogenic disorders requires further study. Prenatal diagnosis is able to provide fetal samples for genetic analysis. Normal results reassure parents that their baby will be unaffected. However, abnormal results gives the couple a difficult decision as whether to ter minate or continue the pregnancy and prepare for postnatal affected infant [26]. In addition, irrespective of the results, some pregnancies may miscarry following the procedures [23, 24, 25].
Pre-implantation genetic testing (PGT) [27] is an alternative to traditional PND allowing the parents a chance to initiate a pregnancy with the confidence that the baby will be un-affected. However, the extremely large size of the DMD gene and the diversity of mutations make molecular genetic testing difficult and labor intensive. Preli minary PGT protocols using single cell multiplex PCR amplifying five dystrophin gene exons in combination with sex identification has been introduced [28]. A similar strategy has been applied in clinical PGT for DMD [29, 30]. PGT for DMD using interphase fluorescence in situ hybridization (FISH) to detect deletions of specific exons within the dystrophin gene has been previously reported [31]. The application of multiple displacement amplification (MDA) prior to PCR has also been applied in PGT of DMD [32]. PGT protocols incorporating the analysis of five or seven exons, four polymorphic markers distributed along the dystrophin gene situated in the two deletion hotspots, and the analysis of amelogenin fragments for sex identification has been described [33]. For PGT-M of DMD, developing PCR protocols for each family are expensive, labor intensive, and time consu ming.
Karyomapping is an advanced molecular method using single nucleotide polymorphism array (aSNP) for simultaneously haplotyping and copy number variation (CNV) analysis [34]. The techniques were first demonstrated using samples from previous PCR based PGT-M of cystic fibrosis in 2009. Karyomapping was also tested on embryo samples of previous PGT cases, including Huntington disease, Peutz-Jeghers syndrome, and Crigler-Najjar syndrome [35]. These procedures gave successful results in 97.7% of samples. However, the studies did not carry out genuine clinical PGT. The first live birth after PGT-M using karyomapping from polar body biopsy was for Smith-Lemli-Opitz syndrome in 2014 [36]. Since the study employed polar body biopsy techniques, post-zygotic CNV cannot be detected. Specifically, the transferred unaffected embryo was not tested by karyomapping. Therefore, the claim that this was the first karyomapping PGT-M live birth was inaccurate. Successful PGT-M using karyomapping for Marfan syndrome was also reported in 2015 [37]. The study performed a SNP on single blastomeres from day 3 embryo biopsy with a SNP demonstrating results in 78–82% and PCR giving results in 87.5% with a healthy male infant as a result. Although there have been several publications regarding PGT-M using karyomapping, most were retrospectively performed from samples of previous PCR based PGT. Only a few had prospective clinical PGT cycles with unimpressive results. Therefore, more details need to be explored regarding the clinical applications of karyomapping.
This study aims to apply a SNP and karyomapping for PGT-M of DMD and PGT-A in 2 clinical PGT cycles in comparison to PCR techniques.
Two families at risk of having an affected DMD offspring joined the project following extensive counselling and the obtaining of informed consent. The project was approved by the Research Ethics Committee of Faculty of Medicine, Chiang Mai University (OBG-2562-06117).
The patient of family DA was 31 years old and her husband was 32 years old. She
and her mother carried DMD c.895G
Both patients underwent IVF procedures with routine ovarian stimulation. Intracytoplasmic sperm injection (ICSI) was performed to avoid sperm DNA conta mination. Blastocysts of good quality were chosen for biopsy with a laser on day 5 post-fertilization. Five trophectoderms were taken for whole genome amplification (WGA) and subsequent molecular testing including e. karyomapping and mutation analysis. After biopsy, all embryos were cryopreserved.
Biopsied trophectoderms were washed thoroughly in phosphate buffered saline
(PBS, Cell Signaling Technology, Theera Trading Co. Ltd. Bangkok, Thailand) with
0.1% polyvinyl alcohol (PVA, Sigma-Aldrich, Chiangmai VM Co., Ltd., Chiang Mai,
Thailand) before transferring to microcentrifuge tubes with a total volume of 4
Amplified MDA samples were tested with SNP array using Illu mina HumanKaryomap-12 DNA Analysis Kit (Bio-Active Co. Ltd., Bangkok, Thailand) according to the manufacturers instructions [34]. SNP genotyping information was analyzed using BlueFuse Multi software version 4.5 (Bio-Active Co. Ltd., San Diego, CA, USA) for karyomapping analysis and molecular cytogenetics. Haplotyping analysis from SNP genotyping information of the couples together with an offspring or an informative relative serving as a reference reveals inheritance of unaffected or affected genes in the embryos allowing the diagnosis of a monogenic disorder of the embryos. Additionally, SNP genotyping provides CNV details of every chromosome. These results were compared with those of PCR.
Mutation analysis was performed to confirm diagnosis results. Aliquots of
amplified WGA products were used for multiplex fluorescent PCR and
mini-sequencing analysis. For family DA, 0.5
Primers | Location on DMD gene | Sequences | Fragment length (bp) | References | |
DMD c.895G |
Exon 11 | forward | 5′-GGCCGGGTTGGTAATATTCT-3′ | 126 | OMIM: NM_004010 |
reverse | 5′-CCTGAGGCATTCCCATCTT-3′ | ||||
DMD c.895G |
Exon 11 | mini-sequencing | 5′-TCAGAAGATGAAGAAACTGAAGTACAA-3′ | ||
5’-5n4 | Intron 4 | forward | 5′-GAAGGGAAAATGATGAATAAAACT-3′ | 134–186 | [40] |
reverse | 5′-GTCAGAACTTTGTCACCTGTC-3′ | ||||
DXS206 | Intron 7 | forward | 5′-TTCTGGTTTTCTGGTCTG-3′ | 218–236 | [41] |
reverse | 5′-GAATCAATCTCTCTGTCAAG-3′ | ||||
DXS1236 | Intron 49 | forward | 5′-GGCAAGTTTCTCTTCGTCGT-3′ | 226–260 | [20] |
reverse | 5′-CGATTCGTGTTTTGCATTGT-3′ | ||||
DXS1214 | Intron 63 | forward | 5′-GCCAGCGTATACCCATTTTG-3′ | 148–162 | [42] |
reverse | 5′- CAGGTAGAAAGATAGCAGGCAAC-3′ | ||||
Amelogenin X/Y | forward | 5′-GCTTGAGGCCAACCATCAG-3′ | X 119 | [39] | |
reverse | 5′-CCTGGGCTCTGTAAAGAATAG-3′ | Y 125 |
A mixture of 1
Mini-sequencing techniques were employed for mutation analysis of DMD
c.895G
A mixture of 1
Two clinical PGT-M cycles for DMD were performed. Nine embryos with good
morphology from each patient were chosen for PGT-M using a SNP with karyomapping
analysis. DNA of the mother of the patient who is a carrier was employed as the
reference. Karyomapping results of family DA (DMD c.895G
Family DA’s haploblock chart of karyomapping analysis.
Happloblock chart of DMD c895G
Embryo No. | Mini-Sequencing DMDc895G |
STR1 | STR2 | STR3 | STR4 | AMXY | Gender | PCR results | Karyomapping analysis | Chromosome analysis | Karyomapping results | Conclusion results | Notes | |||||||
Intron4 | Intron7 | Intron49 | Intron63 | |||||||||||||||||
5’-5n4 | DXS206 | DXS1236 | DXS1214 | |||||||||||||||||
Alleles | Alleles | Alleles | Alleles | Alleles | Alleles | |||||||||||||||
Normal | Mutant | Pat |
Mat |
Pat | Mat | Pat | Mat | Pat | Mat | 119X | 125Y | Pat | Mat | |||||||
Father | G | - | 156 | 232 | 238 | 156 | 119 | 125 | Male | Normal | P1 | Normal | ||||||||
Mother | G | T | 134 | 148 | 228 | 228 | 254 | 232 | 162 | 152 | 119 | - | Female | Carrier | M1/M2 | Carrier | ||||
Mother’s mum | G | T | 150 | 148 | 234 | 228 | 252 | 232 | 160 | 152 | 119 | - | Female | Carrier | M1/M3 | Carrier | ||||
DA1 | G | T | 156 | 148 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Carrier | M1 | 46,XX | Carrier | Carrier | ||
DA2 | G | - | 156 | 134 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Normal | M1/M2 | 46,XX | Carrier or Normal | Ambiguous |
||
DA3 | - | T | - | 148 | - | 228 | - | 232 | - | 152 | 119 | 125 | Male | Affected | M1 | 46,XY | Affected | Affected | ||
DA4 | G | - | - | 134 | - | 228 | - | 254 | - | 162 | 119 | 125 | Male | Normal | M2 | 46,XY | Normal | Normal | ||
DA5 | - | T | - | 148 | - | 228 | - | 232 | - | 152 | 119 | 125 | Male | Affected | M1 | 46,XY | Affected | Affected | ||
DA6 | G | - | 156 | 134 | 232 | 228 | 238 | 254 | 156 | 162 | 119 | - | Female | Normal | M2 | 46,XX | Normal | Normal | ||
DA7 | G | - | - | 134 | - | 228 | - | 254 | - | 162 | 119 | 125 | Male | Normal | M2 | 46,XY | Normal | Normal | ||
DA8 | G | T | 156 | 148 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Carrier | M1/M2 | 46,XX | Carrier or Normal | Ambiguous |
||
DA9 | G | T | 156 | 148 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Carrier | M1 | 46,XX | Carrier | Carrier | ||
DA10 | - | T | - | 148 | - | 228 | - | 232 | - | 152 | 119 | 125 | Male | Affected | Affected | |||||
DA11 | G | T | 156 | 148 | 232 | 228 | - | 232 | - | 152 | 119 | - | Female | Carrier | Carrier | |||||
DA12 | G | T | 156 | 148 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Carrier | Carrier | |||||
DA13 | G | T | 156 | 148 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Carrier | Carrier | |||||
DA14 | G | - | - | 134 | - | 228 | - | 254 | - | 162 | 119 | 125 | Male | Normal | Normal | |||||
DA15 | G | T | 156 | 148 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Carrier | Carrier | |||||
DA16 | G | - | 156 | 134 | 232 | 228 | 238 | 254 | 156 | 162 | 119 | - | Female | Normal | Normal | |||||
DA17 | G | T | 156 | 148 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Carrier | Carrier | |||||
DA18 | G | - | 156 | 134 | 232 | 228 | 238 | 254 | 156 | 162 | 119 | - | Female | Normal | Normal | |||||
DA19 | - | T | - | 148 | - | 228 | - | 232 | - | 152 | 119 | 125 | Male | Affected | Affected | |||||
DA20 | - | T | - | 148 | - | 228 | - | 232 | - | 152 | 119 | 125 | Male | Affected | Affected | |||||
DA21 | G | T | 156 | 148 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Carrier | Carrier | |||||
DA22 | G | - | - | 134 | - | 228 | - | 254 | - | 162 | 119 | 125 | Male | Normal | Normal | |||||
DA23 | G | T | 156 | 148 | 232 | 228 | 238 | 232 | 156 | 152 | 119 | - | Female | Carrier | Carrier | |||||
DA24 | G | - | 156 | 134 | 232 | 228 | 238 | 254 | 156 | 162 | 119 | - | Female | Normal | Normal | |||||
*Pat, Paternal; *Mat, Maternal. |
According to CNV information from karyomapping, three normal (two male, embryos No. DA4 and DA7 and one female, embryo No. DA6) and two carrier (both female, embryos No. DA1 and DA9) embryos with chromosomal balance were fulfilled for transfer (Table 2). During the first embryo transfer, one normal female embryo (embryo No. DA6) was chosen with no resulting pregnancy. In the second transfer, one normal male embryo (embryo No. DA4) was transferred with a resulting normal male infant. Considering that PND procedures result in increased miscarriage rate, the patient refused PND. Postnatal DNA analysis confirmed the PGT results.
Nine embryos of family DB were chosen for PGT-M using a SNP with karyomapping analysis. DNA of the affected son was employed as the reference. Karyomapping results of family DB (DMD exon 8–9 duplication) revealed four normal (embryos No. DB3, DB5, DB8 and DB9), two carriers (embryos No. DB4 and DB7), two affected (embryos No. DB1 and DB2) and one with intragenic recombination (embryo No. 6) (Fig. 2). Microsatellites-based linkage analysis confirmed haplotyping results in all embryos. Additionally, cytogenetic analysis from SNP information demonstrated one normal embryo chromosomally unbalanced, i.e., 45,XX, +2P, –22 (embryo No. DB3) (Fig. 3a) and one normal embryo with uniparental disomy of every chromosome (UPD, embryo No. DB9) (Fig. 3b). Both sets of the chromosomes were maternal, ie unimaternal disomy. Therefore, two normal (both female, embryos No. DB5 and DB8) and two carrier (both female, embryos No. DB4 and DB7) embryos that were chromosomally balanced were fulfilled for transfer (Table 3). All are being cryopreserved for later embryo transfer. Polymorphic marker analysis revealed the absence of extraneous DNA contamination.
Family DB’s haploblock chart of karyomapping analysis. Haploblock chart of DMD exon 8–9 duplication from karyomapping (BlueFuse Multi software) using SNP array information (Illumina HumanKaryomap-12 DNA Analysis Kit) and multiplex fluorescent PCR (F-PCR) for the couples at risk of having DMD exon 8–9 duplication offspring (family DB). DNA of the affected son was employed as the reference. Haplotyping of DMD gene was demonstrated together with linkage analysis of short tandemly repeats (STR) and chromosome analysis results.
Family DB’s chromosome balance analysis from aSNP results. (a) shows copy number variation (CNV) of embryo DB3, 45,XX, +2p, –22. (b) shows CNV of embryo DB9, 46,XX, uniparental disomy of every chromosome.
Embryo No. | STR1 Intron 4 | STR2 Intron 7 | STR3 Intron 49 | STR4 Intron 63 | AMXY | Gender | PCR results | Karyomapping analysis | Chromosome analysis | Karyomapping results | Conclusion results | Notes | ||||||
Alleles | Alleles | Alleles | Alleles | Alleles | ||||||||||||||
Pat |
Mat |
Pat | Mat | Pat | Mat | Pat | Mat | 119 | 125 | Pat | Mat | |||||||
Father | 156 | 232 | 254 | 156 | 119 | 125 | Male | Normal | - | Normal | ||||||||
Mother | 148 | 152 | 228 | 226 | 234 | 238 | 156 | 158 | 119 | - | Female | Carrier | M1/M2 | Carrier | ||||
Son | - | 152 | - | 226 | - | 238 | - | 158 | 119 | 125 | Male | Affected | - | M1 | Affected | |||
DB1 | - | 152 | - | 226 | - | 238 | - | 158 | 119 | 125 | Male | Affected | M1 | 46,XY | Affected | Affected | ||
DB2 | - | 152 | - | 226 | - | 238 | - | 158 | 119 | 125 | Male | Affected | M1 | 46,XY | Affected | Affected | ||
DB3 | 156 | 148 | 232 | 228 | 254 | 234 | 156 | 156 | 119 | - | Female | Normal | M2 | 45,XX | Normal | Normal |
||
+2p,-22 | ||||||||||||||||||
DB4 | 156 | 152 | 232 | 226 | 254 | 238 | 156 | 158 | 119 | - | Female | Carrier | M1 | 46,XX | Carrier | Carrier | ||
DB5 | 156 | 148 | 232 | 228 | 254 | 234 | 156 | 156 | 119 | - | Female | Normal | M2 | 46,XX | Normal | Normal | ||
DB6 | 156 | 152 | 232 | 226 | 254 | 238 | 156 | 158 | 119 | - | Female | Carrier | M2/M1 | 46,XX | Carrier | Ambiguous | Intragenic recombination | |
DB7 | 156 | 152 | 232 | 226 | 254 | 238 | 156 | 158 | 119 | - | Female | Carrier | M1 | 46,XX | Carrier | Carrier | ||
DB8 | 156 | 148 | 232 | 228 | 254 | 234 | 156 | 156 | 119 | - | Female | Normal | M2 | 46,XX | Normal | Normal | ||
DA9 | 156 | 148 | 232 | 228 | 254 | 234 | 156 | 156 | 119 | - | Female | Normal | M2 | 46,XX UPD | Normal | Normal |
||
*Pat, Paternal; *Mat, Maternal. |
In this study, high resolution SNP array provided haplotyping-based diagnosis of
DMD in 2 clinical PGT-M cycles. Karyomapping results were verified by PCR-based
analysis. Novel multiplex PCR incorporating with mini-sequencing and fluorescent
PCR was developed for DMD c895G
PCR can analyze as many embryos as needed with lower extra expense, while high additional expense is major concern for microarray and prevents some embryos from analysis as in family DA’s. In family DA, from all 24 embryos with good quality, only nine were analyzed by karyomapping, while all 24 were analyzed by PCR. One kit of Illu mina HumanKaryomap-12 DNA Analysis Kit can analyzed 12 samples at a time, sparing three samples for the parents and one close relative as references, it is possible to analyze up to 9 embryos in one kit. Analyzing more embryos will double or triple the cost of diagnosis. Therefore, karyomapping results were available for only 9 embryos, while PCR results were presented for all embryos of family DA in this study (Table 2).
The advantages of karyomapping over the conventional PCR-based diagnosis include the diagnosis of duplication or large insertion (i.e., family DB’s) and deletion mutations with unknown breakpoints, rescuing PCR results with ADO and the additional CNV information. Due to the advantage of SNP, origins of chromosomal gain and loss and UPD can be revealed. In this study, embryos No. DB3 and DB9 were found to be 45,XX, +2p, –22 and uni-maternal disomy, respectively. All gained and lost chromosomes belonged to the mother. It was demonstrated that karyomapping can analyze both point mutation (family DA) and large duplication (family DB) without the need to have protocol modifications. Therefore, karyomapping is a widely applicable PGT-M protocol.
DMD is an X-linked recessive disorder, therefore, male offspring can be either normal or affected, while female offspring can be either normal, carrier or affected. In this study, the mothers of both families were carriers of mutant genes, but the fathers did not carry any mutant genes. Therefore, PGT-M for DMD focused on maternal DMD alleles as there was no need to analyze paternal allele.
Interestingly, when the pathogenic variant is known, karyomapping can be used independently for PGT because aCGH can identify any recombination events eli minating any misdiagnosis from recombination. In families with a history of muscular dystrophy with no known pathogenic variants, a third are new mutations where the mothers’ pathogenic variant is not known. Pprocedures of PGT for these families are very challenging. Types of particular muscular dystrophy need to be confirmed by clinical geneticists using clinical criteria, biochemical assays and histology. When the particular type of muscular dystrophy was confirmed, whole exome sequencing (WES) is performed in the members of the family including probands. Candidate genes for the particular type of muscular dystrophy are focused. Bioinformatics can be helpful in accelerating mutation identification process with PGT then being performed for the families.
The strength of this study is the ability of performed karyomapping along with
PCR analysis for PGT of DMD in two families. Karyomapping provided haplotyping
based diagnosis and chromosome balance information of the embryos. PCR revealed
direct mutation analysis (family DA) and microsatellites based linkage analysis
(both families) results. DMD families in this study possessed single nucleotide
mutation (c895G
Two clinical PGT-M cycles using karyomapping were performed for both families at
risk of having DMD (c895G
ADO, allele drop out; aSNP, single nucleotide polymorphism microarray; CNV, copy number variation; DMD, Duchenne muscular dystrophy; ICSI, intracytoplasmic sperm injection; IVF, in vitro fertilization; MDA, multiple displacement amplification; PCR, polymerase chain reaction; PGT-A, pre-implantation genetic testing for aneuploidy; PGT-M, pre-implantation genetic testing for monogenic disorders; PND, prenatal diagnosis; WGA, whole genome amplification.
Study conception and design were performed by SM, SP and WP. Ovarian stimulation, oocytes collection and embryology laboratory were performed by TP and SM. aSNP and karyomapping analysis were performed by RS, SM, SP and WP. PCR analysis and mini-sequencing were performed by WS, SP and WP. Data collection and analysis were performed by SP, SM and WP. CT took care of clinical diagnosis and assessment. Prenatal and postnatal diagnosis were performed by TT and WP. SM and SP contributed equally to this work. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Informed consent of all participants were obtained. The project was approved by the Research Ethics Committee of Faculty of Medicine, Chiang Mai University (OBG-2562-06117).
Thanks to all the peer reviewers for their opinions and suggestions.
This study was supported by the National Research Council of Thailand and Chiang Mai University Research Fund (No. CMU-2563).
The authors declare no conflict of interest.