SEARCH

SEARCH BY CITATION

Keywords:

  • array CGH;
  • prenatal diagnosis;
  • ultrasound abnormalities

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. REFERENCES

Objective

Array comparative genomic hybridization (CGH) is transforming clinical cytogenetics with its ability to interrogate the human genome at increasingly high resolution. The aim of this study was to determine whether array CGH testing in the prenatal population provides diagnostic information over conventional karyotyping.

Methods

MEDLINE (1970 to December 2009), EMBASE (1980 to December 2009) and CINAHL (1982 to December 2009) databases were searched electronically. Studies were selected if array CGH was used on prenatal samples or if array CGH was used on postnatal samples following termination of pregnancy for structural abnormalities that were detected on an ultrasound scan. Of the 135 potential articles, 10 were included in this systematic review and eight were included in the meta-analysis. The pooled rate of extra information detected by array CGH when the prenatal karyotype was normal was meta-analyzed using a random-effects model. The pooled rate of receiving an array CGH result of unknown significance was also meta-analyzed.

Results

Array CGH detected 3.6% (95% CI, 1.5–8.5) additional genomic imbalances when conventional karyo-typing was ‘normal’, regardless of referral indication. This increased to 5.2% (95% CI, 1.9–13.9) more than karyotyping when the referral indication was a structural malformation on ultrasound.

Conclusions

There appears to be an increased detection rate of chromosomal imbalances, compared with conventional karyotyping, when array CGH techniques are employed in the prenatal population. However, some are copy number imbalances that are not clinically significant. This carries implications for prenatal counseling and maternal anxiety. Copyright © 2010 ISUOG. Published by John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. REFERENCES

Structural chromosomal anomalies are a major cause of perinatal morbidity and mortality1–3. Since the 1960s, full chromosome analysis has been the mainstay of diagnosing karyotypic abnormality. Despite improvements in cytogenetic resolution, karyotyping may only detect anomalies to a resolution of 5–10 Mb4. Array comparative genomic hybridization (CGH) has several advantages over conventional chromosome analysis. A high-resolution oligonucleotide array is capable of detecting changes to a resolution of 1 kb (smaller than the average gene), and customized arrays designed for prenatal diagnosis have been developed.

One of the current challenges of the application of CGH microarrays in the clinical setting is determining whether a copy number imbalance is de novo and likely to be causative, or inherited and likely to be benign. When analyzing array data in a clinical setting, clinical cytogeneticists categorize copy number variants (CNVs) into those that are likely to be ‘benign’, those that are likely to be ‘pathogenic’ and those of ‘unknown clinical significance’. CNVs that overlap critical regions of established microdeletion or microduplication syndromes are likely to be pathogenic5.

Prenatal fetal karyotyping is offered during a pregnancy because a screening test has indicated a ‘high’ risk that the fetus may have aneuploidy, because of a structural anomaly on ultrasound examination, because of a family history of chromosomal abnormality or because of parental choice. Conventional chromosome analysis using G-banding will detect chromosome anomalies such as trisomies 21, 18 and 13, and monosomy X, along with many structural rearrangements. However, there are a small number of pregnancies in which conventional karyotyping is reassuring but uncertainties persist about the underlying etiology.

Studies have been published indicating that array technologies may detect previously ‘undiagnosed’ chromosome aberrations. These studies have also highlighted limitations to this technology, not least the identification of unknown CNVs that may increase parental anxiety and difficulties in counseling. As this technology will become increasingly important in prenatal diagnosis, we sought to systematically review the literature and describe the evidence that array CGH may offer diagnostic information over conventional karyotyping, and specifically the value of arrays, in addition to karyotyping, for fetuses found to have a structural abnormality following an ultrasound scan.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. REFERENCES

Our systematic review followed a prospective protocol developed using widely recommended and comprehensive methodology6.

Data sources

The search focused only on prenatal studies using microarray technology. A search strategy was developed based on existing advice for prevalence searches7, 8. MEDLINE (1970 to December 2009), EMBASE (1980 to December 2009) and CINAHL (1982 to December 2009) databases were searched electronically. The search of MEDLINE and EMBASE captured citations containing the relevant MeSH keywords and word variants for ‘microarray’ and ‘prenatal’. The following terms were used to describe microarrays: array comparative genomic hybridization, microarray and oligonucleotide array. Similarly, antenatal diagnostics, fetal diagnostics, prenatal and fetal were used to capture ‘prenatal’. Bibliographies of relevant articles were manually searched to identify papers not captured by the electronic searches. Web of Science (1996–2009) was used to capture any gray literature. Experts were also contacted for completeness of the search (the authors of the papers Coppinger et al.9 and Van den Veyver et al.10). There were no language restrictions in the search or selection of papers.

Eligibility criteria for selecting studies

Studies were selected in a two-stage process. Initially, all abstracts or titles in the electronic searches were scrutinized by two reviewers (S.H. and S.P.) and full manuscripts of potentially eligible citations were obtained. Differences were resolved by discussion. Unresolved disagreements were resolved by a third reviewer (A.C.). Studies were selected if array CGH or microarray had been used on prenatal specimens (analyzed either during pregnancy or after delivery). We also selected reports if the same technology had been used on postnatal specimens following termination of pregnancy for structural abnormalities detected on an ultrasound scan. Papers were excluded if the testing was performed postnatally and the indication for running the array had not been determined prenatally. Papers were also excluded if the array was performed on children or adults or if it was used for preimplantation genetic diagnosis or for the diagnostic investigation of recurrent miscarriages. Finally, papers were excluded if they used the CGH technique and not array CGH. Non-English studies were assessed by people with command of the relevant language if the title or abstract appeared to fit the criteria. Only papers that allowed generation of a 2 × 2 table (comparing karyotyping with array) were included. In two instances, in order to construct a 2 × 2 table the authors were contacted. In the case of Coppinger et al.9, direct discussion/correspondence allowed extra information to be obtained and for us to include the paper. In the second case (Van den Veyver et al.10), we were still unable to include the paper because the authors were unable to provide us with enough information to complete a 2 × 2 table.

Data extraction

Data were extracted by two reviewers (S.H. and S.P.). For each of the outcomes, data were extracted into tables, giving descriptive and numerical information for each study. Data were extracted on study characteristics and data quality. Data were used to construct 2 × 2 tables of test accuracy comparing normal and abnormal karyotype results against normal and abnormal microarray results. Case studies of fewer than five cases were excluded from the meta-analysis.

Quality assessment

All articles meeting the selection criteria were assessed for quality using items from validated tools (Figure 1). A study was considered to be of good quality if it used a prospective design, if it used a representative population (i.e. it used array technology on all samples, not just on those with an exclusively normal or abnormal conventional karyotype), if it performed array testing on parents to aid interpretation of CNVs and if it used a validated assessment tool (i.e. an identifiable, reproducible array). Expert opinion from cytogeneticists (D.M. and E.V.D.) at the West Midlands Regional Genetics Department was sought to determine the validity of the array used.

thumbnail image

Figure 1. Quality assessment of the 10 papers included in the systematic review.

Download figure to PowerPoint

Data synthesis

The analysis was performed in two steps depending on the way in which the CNVs were grouped. The first analysis grouped pathogenic, unknown and benign CNVs as array-detectable variants. The second analysis moved benign CNVs into the normal group, leaving the abnormal CNVs as those that were pathogenic plus those that were of unknown significance and potentially pathogenic. Analysis was performed for samples undergoing both karyotyping and array, regardless of the clinical indication (maternal anxiety, high risk on screening serum for Down syndrome and/or structural abnormality on ultrasound scan). Analysis was then performed for those undergoing chromosomal analysis because of a structural abnormality found during an ultrasound scan.

We explored the possibility of separating the data further by attempting to compare different sample types (e.g. from amniocentesis and chorionic villus sampling) and the effect that this may have on the array CGH results. However, this was not possible because, with the exception of one paper9, the studies did not record from which sample type the DNA was extracted when an abnormal chromosomal result was found.

Using 2 × 2 tables, we computed and pooled the percentage agreement between the two technologies (with 95% CI) for the articles overall. The calculated percentage of extra cases identified by array in those with a normal karyotype (both overall and by referral indication) with 95% CI was calculated and pooled. Finally, we calculated and pooled the percentage of cases in which a result of ‘unknown significance’ was reported. Heterogeneity in rates was examined graphically and statistically. For graphical assessment, Forest plots of point estimates of rates and their 95% CI were used. For exploration of reasons for heterogeneity, stratified analysis was performed according to the features of the population (indication for referral). For the meta-analysis, log rates were pooled, weighting each study by the inverse of its variance11, and the summary estimates were exponentiated. A random-effects model was used in the light of heterogeneity. All statistical analyses were performed using Stata 8.0 statistical software (Stata Corp., College Station, TX, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. REFERENCES

The process of literature identification and selection is summarized in Figure 2. There were 10 primary articles identified as meeting the selection criteria9, 12–20. Of the original articles, 97 were excluded because they did not meet the selection criteria. The remaining 38 articles were obtained and reviewed, and a further eight articles were requested after reviewing the reference lists. Of these 46 articles, 36 were then excluded as they did not meet selection criteria. The 10 primary studies, containing 798 participants, met the inclusion criteria to be included in the systematic review9, 12–20. Eight of these studies were included in the meta-analysis9, 12–14, 17–20 and two were excluded from the meta-analysis because array CGH was only performed when an abnormal karyotype had been detected15, 16.

thumbnail image

Figure 2. Selection process of the 10 papers included in the systematic review.

Download figure to PowerPoint

Table 1 summarizes the characteristics of the publications used, including the design of the study (retrospective or prospective), the array type, sample type, if the array was targeted or covered the whole genome, the indication for the array (divided into different structural abnormalities where known) and the sample size. Study quality assessment showed deficiencies in many areas of methods (Figure 1). Only three papers met all four quality criteria: being prospective in design; using array and karyotyping on a representative population (i.e. the population did not have all known abnormal/normal karyotypes); investigating parents to aid interpretation of CNVs; and using a validated assessment tool12, 13, 20.

Table 1. Characteristics of the studies included
Reference/ study designArray typeProbeGenome (WG/TG)Sample typeIndication for arraySample size (n)
  1. BAC, bacterial artificial chromosome; BCM, Baylor College of Medicine; CH, cystic hygroma; CNS, central nervous system; CNVs, copy number variants; CVS, chorionic villus sampling; GI, gastrointestinal; IUGR, intrauterine growth restriction; MCA, middle cerebral artery; NF, nuchal fold; NT, nuchal translucency; PAC, plasmid artificial chromosome; SNP, single nucleotide polymorphism; SUA, single umbilical artery; TG, targeted genome; WG, whole genome.

Coppinger et al.Signature prenatal chip V> 2100 BAC probesTGAmniotic fluid n = 40Anxiety n = 662
 (2009)9 4.0, commercial  CVS n = 22Family history n = 19 
 Prospective    Advanced maternal age n = 3 
     Abnormal maternal serum screen n = 1 
     Abnormal ultrasound n = 33 
      (CNS n = 3; skeletal n = 1; CH/NT/hydrops n = 18; GI n = 1; IUGR n = 1; MCA n = 7; 
      unspecified n = 7) 
     All normal karyotype n = 62 
 Signature chip whole4670 BAC probesWGAmniotic fluid n = 149Anxiety n = 3182
  genome, commercial  CVS n = 30Family history n = 17 
    Unspecified prenatal cellAdvanced maternal age n = 5 
     type n = 3Abnormal maternal serum screen n = 2 
     Abnormal ultrasound n = 155 
      (CNS n = 16; musculoskeletal n = 8; 
      cleft lip n = 6; CH/NT/hydrops 
      n = 24; GI n = 4; renal/ambiguous 
      genitalia n = 4; cardiac n = 13; 
      IUGR n = 7; MCA n = 63; 
      unspecified n = 10) 
Kleeman et al.Signature prenatal chip V1887 BAC probesTGAmniotic fluid n = 47Abnormal ultrasound scan and50
 (2009)19 4.0, commercial, 26 patients    normal karyotyping n = 50 (cardiac n = 24; 
 ProspectiveSignature whole-genome chip,4685 BAC probesWGCVS n = 3 CNS n = 6; skeletal n = 6; 
  commercial    urogenital n = 4; cleft lip/palate 
      n = 2; CH/NT/hydrops n = 3; 
      GI n = 2; multiple n = 17; growth 
      disorder n = 3) 
Tyreman et al.Genechip SNP 6.0 array946 000 probes (CNVs)WGAmniotic fluid n = 87Ultrasound anomaly n = 106106
 (2009)18 (Affymetrrix), commercial  CVS n = 15 (cardiac n = 34; multisystem n = 24; 
 Retrospective   Placenta biopsy following large NT/hydrops/CH n = 18; 
     termination of CNS n = 16; skeletal n = 6; 
     pregnancy n = 4 abdominal wall n = 2; others n = 6) 
Bi et al.BCM V6 Oligonucleotide44 000 oligonucleotideWG and TGAmniotic fluidMaternal age n = 614 pregnancies
 (2008)13 array (V6 Oligo) probes  Anomaly on ultrasound n = 515 fetuses
 Prospective (Agilent), commercial   Family history abnormality n = 2 
     Multiple miscarriages n = 1 
Shaffer et al.Prenatal BAC array2100 BAC probesTGPrenatal cultured amnioticFamily history n = 19151
 (2008)14 version (signature),  fluid or CVS; postnatalMaternal age n = 2 
 Prospective commercial   bloodParental anxiety n = 20 
     Anomaly on ultrasound n = 110 
      (abnormal genitalia n = 6; CNS 
      n = 19; CH/NF/NT/hydrops n = 22; 
      GI n = 9; dysmorphic n = 4; 
      cardiac n = 17; IUGR n = 2; 
      multiple n = 3; micrognathia n = 2; 
      midline defect n = 8; renal n = 2; 
      skeletal n = 11; SUA n = 1; others n = 4) 
Vialard et al.Genosensor BAC/PAC287 BAC probesTGMuscle biopsy n = 15;Two or more abnormalities39
 (2008)20 array 300   lung biopsy n = 13; thymus cardiovascular/urogenital/ 
 Prospective (Vysis/Abbott),   n = 4; skin n = 3; liver n = 3; skeletal/digestive/CNS 
  commercial   bladder n = 1  
De Gregori et al.60-mer oligonucleotide60-mer oligonucleotideWGNot statedReciprocal translocations n = 1417
 (2007)16 microarray, commercial probes  Maternal age n = 3 
 Retrospective      
Rickman et al.BAC/PAC resolution600 BAC probesTGCultured amniocytes or CVSPreviously known karyotypes30
 (2006)15 10 Mb, common   All known unbalanced 
 Retrospective microdeletion syndrome, own array    rearrangements 
Sahoo et al.BCM v4.0, Baylor,366 BAC probesTGAmniotic fluid n = 56Increased maternal age98
 (2006)12 commercial  (26 uncultured)Increased serum screen 
 Prospective   CVS n = 42 (32 uncultured)Family history 
     Abnormality on ultrasound 
Le Caignec et al.Genosensor BAC/PAC287 BAC probesTGFrozen fetal tissueAll normal karyotype49
 (2005)17 array 300   All had malformations 
 Retrospective (Vysis/Abbott),   All had at least three anomalies in 
  commercial    CVS/urogenital/digestive/CNS 

The overall agreement between karyotype and array results was 88.2% (95% CI, 79.2–98.2). When benign CNVs were removed from the abnormal array group and treated as normal array results (as described above in the second analysis) the agreement was increased, to 95.6% (95% CI, 86–100%). The data were homogeneous (P = 0.99 and P = 0.97, respectively). Four out of 10 papers were used to review the overall agreement between karyotype results and array results (n = 333). Six out of 10 papers14–19 could not be included because the dataset was not complete (i.e. the sample population was skewed by only using arrays on those samples with all normal or all abnormal karyotypes).

Array CGH detected 12% (95% CI, 8.8–16.4%) more chromosomal imbalances overall when karyotyping was ‘normal’ (Figure 3a) when the array was performed for any clinical indication. When benign CNVs were recognized, removed and treated as normal results, the detection rate decreased to 3.6% (95% CI, 1.5–8.5%) (Figure 3b). This 3.6% included all CNVs known to be pathogenic and those of unknown significance with the potential to be pathogenic. We therefore calculated how often a result of unknown significance would be found when array CGH was performed prenatally for any clinical indication. Results of ‘unknown significance’ were found in 1.1% (95% CI, 0.4–2.7%) of cases. Eight out of 10 papers were used for these meta-analyses9, 12–14, 17–20 (n = 751). Two were excluded because they did not contain data for chromosomal anomaly detection rate by array when a normal karyotype was reported15, 16. The paper by Sahoo et al.12 was included in the analysis as having three results of unknown significance; the table of results in this paper (Table 1) implies that one of these (Case 3) was probably benign, but the text describes how the parents were ‘counseled as to the unknown significance of the results’ and it is therefore included as a result of unknown significance. These data were heterogeneous.

thumbnail image

Figure 3. Forest Plot. Meta-analysis of the detection rate of chromosomal imbalances by array comparative genomic hybridization when karyotyping is normal and chromosomal testing is performed for either any clinical indication or an abnormal ultrasound scan. (a) Analysis 1: array results showing chromosomal imbalances that are copy number variants (CNVs) which are pathogenic, of unknown significance or benign. (b) Analysis 2: array results showing chromosomal differences that are CNVs which are pathogenic or of unknown significance. Benign CNVs were removed and treated as normal results. TG, targeted genome; WG, whole genome.

Download figure to PowerPoint

One of the papers appeared to contribute disproportionately to the heterogeneity of the data18. This paper used a higher resolution array (Affymetrix SNP 6) and did not use parental testing for clarification of CNVs of unknown significance. It therefore had a high detection rate of all CNVs: pathological, unknown significance and benign. Therefore, we performed a sensitivity analysis by excluding the results from this paper. With this paper excluded, the array detected 10.7% (95% CI, 9.1–12.6) more chromosomal imbalances when karyotyping was normal compared with 12% (95% CI, 8.8–16.4%) with the results of this paper included. With exclusion of the paper by Tyreman et al.18, and when benign CNVs were removed and treated as normal results (analysis 2), the detection rate of chromosomal imbalances by array CGH decreased from 3.6% to 2.9% (95% CI, 1.3–6.3) when karyotyping was normal. Exclusion of this paper did not significantly reduce the array CGH detection rate; in addition, the data were still heterogeneous. Taking this into account, and given that the paper by Tyreman et al.18 is an important paper with one of the largest cohorts of patients, we feel that its inclusion is important to present the totality of evidence.

Conventional karyotyping did not detect any chromosomal imbalances that were not detected by array CGH in the same eight papers used in the meta-analysis described above. Two papers included in the systematic review15, 16, but not in the meta-analysis, used array CGH only when an abnormal karyotype had been found. They found that array CGH was not able to detect one case of triploidy and 14 cases of balanced translocation. These two papers were not included in the meta-analysis because they only studied cases with abnormal karyotype results and therefore a 2 × 2 table of their results could not be formed.

Results in cases when a structural abnormality was noted on ultrasound scan

Array technology detected overall 11.2% (95% CI, 5.7–22.1) more chromosomal imbalances above that of conventional karyotyping (Figure 3a). When ‘benign CNVs’ were removed from the analysis and placed with the normal array results, the detection rate of chromosomal abnormalities decreased to 5.2% (95% CI, 1.9–13.9); this included results that are known to be pathological, and those of unknown significance with the potential to be pathological (Figure 3b). Six papers were used13, 14, 17–20. These papers all contained results on patients who had undergone karyotyping and array tests because they had pregnancies where a structural fetal malformation was suspected on ultrasound scanning (n = 359). The size of these chromosomal imbalances ranged, depending on the resolution of the array CGH used, from 60 kb18 to 60 Mb17. In 1.9% (95% CI, 0.4–9.5) of cases where the patient was referred because of a fetal anomaly on an ultrasound scan and where conventional karyotyping was ‘normal’, a result of ‘unknown significance’ was reported. These data were heterogeneous (P < 0.0001).

The actual numbers of different structural abnormalities were recorded in four out of the six papers9, 14, 18, 19. Where possible, the different structural abnormalities are recorded in Table 1. The pooled data from the four papers show that the largest numbers of patients had cardiac abnormalities (n = 88), increased nuchal translucencies, cystic hygromata or hydrops (n = 82), or central nervous system abnormalities (n = 60). It was not possible to divide the structural abnormalities into groups to perform separate analysis for array CGH testing because all four papers included groups with either ‘multiple congenital abnormalities’, which was not broken down further, or ‘other’, where the abnormality was not specified. None of the papers included single ‘soft’ markers, although the paper by Tyreman et al. included multiple soft markers18. The vast majority were major abnormalities. The papers also did not always provide data postdelivery/postmortem so it was not possible in most cases to establish if the abnormality was confirmed or, in the case of increased nuchal translucency, if it was linked to other structural abnormalities.

Conventional karyotyping did not detect any chromosomal imbalances that were not detected also by array CGH.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. REFERENCES

Good overall agreement between prenatal array technology results and conventional cytogenetic karyotyping from data in the current literature was obtained, as expected. This increased further with the removal of ‘benign’ CNVs, identified by the examination of databases of DNA made available by collaborations such as the Wellcome Trust Case Control Consortium21 or the Database of Genomic Variants (DGV)22. It is to be expected that the percentage agreement between technologies would increase with the removal of ‘benign’ CNVs because the remaining microarray results are more likely to be pathogenic.

Array CGH detected notably more chromosomal imbalances than conventional karyotyping, both when it was performed prenatally for any clinical indication and when it was performed for a structural abnormality detected during an ultrasound scan. As well as results of a known pathogenic nature, this also includes results of unknown pathogenic significance.

Although only occurring rarely, these results are of concern as they may both increase parental anxiety and lead to problems in prenatal counseling. The frequency of CNV of unknown significance is increased if parental samples are not available as analysis will reveal whether the variant detected in the fetus is familial or de novo. However, because of incomplete/variable penetrance, familial variants are not always benign5.

Karyotyping did not detect any chromosomal imbalances when array CGH was reported as normal in those papers used in the meta-analysis9, 12–14, 17–20. This is reassuring because it suggests that array CGH alone will not miss many significant results, and that the areas in which microarray technology is weak (balanced translocations, inversions and a lower sensitivity for triploidy) are not common in a typical referral population.

The strength of the study lay in the rigor of the methodology. It met the quality criteria laid down in the MOOSE23 statement. The meta-analysis contained a relatively small sample size of 751 participants for overall analysis and 409 participants with fetal anomalies identified using ultrasound. However, as many studies published in this area are single case studies, this would appear to be a fairly large cohort. Papers were heterogeneous, with both prospective and retrospective methodology, different indication for referral and different microarrays used. This accounts for the large CIs in our analysis. The inclusion of some published studies that include participants with a known normal karyotype may have influenced the results because they are not necessarily a representative population. Array technologies have increased in resolution since these studies were conducted and they have also become commercially available, allowing greater validation of results between different published studies. The authors cannot allow for ascertainment bias towards cases that clinicians may have felt would have yielded an abnormal result from array and therefore included them in the results. It is recognized that this is a developing scientific field and the literature on this topic is increasing with time. These data represent a critical appraisal of the literature to date and provide a summary for clinicians at the present time.

This systematic review provides evidence of the relative advantage of using array testing in prenatal diagnosis, even when the karyotype is normal. The additional detection of CNVs by array technologies is a combination of known pathological findings but also results of ‘unknown significance’. With emerging array technologies of increasing resolution, the amount of CNVs that are uninterpretable will increase and be associated with increased parental anxiety. Up to 12% of any individual's genome is likely to exhibit normal copy number variation and there is emerging evidence of a huge degree of structural complexity within these chromosomal regions24. In addition, limited data exist on the prevalence of CNVs between different ethnic populations25. These concerns have led to recent recommendations from the American College of Obstetricians and Gynecologists (ACOG) that conventional karyotyping should remain the principal cytogenetic tool in prenatal diagnosis. ACOG has also suggested that targeted arrays can be offered as an adjunct in prenatal cases with abnormal anatomical findings and a normal conventional karyotype26.

Conclusion

There appears to be an increased detection rate of chromosomal abnormalities with array CGH, both for prenatal indications overall and when congenital malformations were noted on ultrasound scans. However, the large CIs obtained in the analysis show that more work is required before we can answer the question of absolute detection rate over conventional karyotyping.

Further prospective research is needed in this area on a large cohort that has undergone both karyotyping and analysis using a commercial reproducible array. The optimum resolution of an array to be used in a prenatal setting has not yet been decided. A targeted array is suitable because of its ability to identify CNVs in known disease-specific loci of the human genome, but risks missing a pathogenic CNV that is absent in these particular genomic regions. A high-resolution array will have the ability to detect a greater number of CNVs but risks having more results of unknown certainty, which need additional time for interpretation and provoke additional uncertainty. Perhaps while this technology is being investigated, targeted arrays are more suitable.

Health economic assessment of microarrays is important when evaluating implementation of this prenatal diagnostic test in routine practice. At present within the UK National Health Service (NHS) setting, almost all samples obtained prenatally are screened using quantitative polymerase chain reaction (Q-PCR) to exclude common trisomies. A proportion of centers also continue to offer full karyotyping and, depending on the type of structural malformation noted, target testing for specific chromosome anomalies is performed.

The cost-effectiveness of these tests depends upon where arrays sit in the cytogenetic testing process. For instance, if Q-PCR continues to be utilized, then the cost of arrays will be relatively expensive compared with the situation where arrays are introduced as the only source of cytogenetic analysis. Also, the resolution of arrays utilized will be important regarding identification of the proportion of CNVs of uncertain significance. This is likely to affect the uptake of genetic counseling, which will adversely affect the economic costs. With the increasing utilization of this technology, the cost of individual arrays is rapidly falling.

Finally there is an urgent requirement for patient satisfaction and qualitative research into the emotional response in parents with the implementation of such prenatal diagnostic techniques.

REFERENCES

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. REFERENCES
  • 1
    Genest DR, Williams MA, Greene MF. Estimating the time of death in stillborn fetuses: I. Histologic evaluation of fetal organs; an autopsy study of 150 stillborns. Obstet Gynecol 1992; 80: 575584.
  • 2
    Genest DR. Estimating the time of death in stillborn fetuses: II. Histologic evaluation of the placenta; a study of 71 stillborns. Obstet Gynecol 1992; 80: 585592.
  • 3
    Genest DR, Singer DB. Estimating the time of death in stillborn fetuses: III. External fetal examination; a study of 86 stillborns. Obstet Gynecol 1992; 80: 593600.
  • 4
    Kirchhoff M, Rose H, Lundsteen C. High resolution comparative genomic hybridisation in clinical cytogenetics. J Med Genet 2001; 38: 740744.
  • 5
    Lee C, Iafrate AJ, Brothman AR. Copy number variations and clinical cytogenetic diagnosis of constitutional disorders. Nat Genet 2007; 39: S48S54.
  • 6
    Khan KS, Ter Riet G, Glanville J, Sowden AJ, Kleijnen J. Undertaking Systematic Reviews of Research on Effectiveness. CRD's Guidance for Carrying Out or Commissioning Reviews (2nd edn). CRD Report No. 4. NHS Centre for Reviews and Dissemination (CRD), University of York: York, 2001.
  • 7
    McKibbon A, Eady A, Marks S. Identifying and selecting studies for inclusion. In PDQ Evidence-Based Principles and Practice. BC Decker Inc.: Hamilton, Canada, 1999; 125127.
  • 8
    Glasziou P, Irwig L, Bain C, Colditz G. Frequency and rate. In Systematic Reviews in Health Care: A practical guide (2nd edn). Cambridge: Cambridge University Press, 2001; 6773.
  • 9
    Coppinger J, Alliman S, Lamb AN, Torchia BS, Bejjani BA, Shaffer LG. Whole-genome microarray analysis in prenatal specimens identifies clinically significant chromosome alterations without increase in results of unclear significance compared to targeted microarray. Prenat Diagn 2009; 29: 11561166.
  • 10
    Van den Veyver IB, Patel A, Shaw CA, Pursley AN, Kang SH, Simovich MJ. Clinical use of array comparative genomic hybridization (aCGH) for prenatal diagnosis in 300 cases. Prenat Diagn 2009; 29: 2939.
  • 11
    Laird NM, Mosteller F. Some statistical methods for combining experimental results. Int J Technol Assess Health Care 1990; 6: 530.
  • 12
    Sahoo T, Cheung SW, Ward P, Darilek S, Patel A, del Gaudio D, Kang SH, Lalani SR, Li J, McAdoo S, Burke A, Shaw CA, Stankiewicz P, Chinault AC, Van den Veyver IB, Roa BB, Beaudet AL, Eng CM. Prenatal diagnosis of chromosomal abnormalities using array-based comparative genomic hybridization. Genet Med 2006; 8: 719727.
  • 13
    Bi W, Breman AM, Venable SF, Eng PA, Sahoo T, Lu XY. Rapid prenatal diagnosis using uncultured amniocytes and oligonucleotide array CGH. Prenat Diagn 2008; 28: 943949.
  • 14
    Shaffer LG, Coppinger J, Alliman S, Torchia BA, Theisen A, Ballif BC, Bejjani BA. Comparison of microarray-based detection rates for cytogenetic abnormalities in prenatal and neonatal specimens. Prenat Diagn 2008; 28: 789795.
  • 15
    Rickman L, Fiegler H, Shaw-Smith C, Nash R, Cirigliano V, Voglino G, Ng BL, Scott C, Whittaker J, Adinolfi M, Carter NP, Bobrow M. Prenatal detection of unbalanced chromosomal rearrangements by array CGH. J Med Genet 2006; 43: 353361.
  • 16
    De Gregori M, Ciccone R, Magini P, Pramparo T, Gimelli S, Messa J, Novara F, Vetro A, Rossi E, Maraschio P, Bonaglia MC, Anichini C, Ferrero GB, Silengo M, Fazzi E, Zatterale A, Fischetto R, Previderé C, Belli S, Turci A, Calabrese G, Bernardi F, Meneghelli E, Riegel M, Rocchi M, Guerneri S, Lalatta F, Zelante L, Romano C, Fichera M, Mattina T, Arrigo G, Zollino M, Giglio S, Lonardo F, Bonfante A, Ferlini A, Cifuentes F, Van Esch H, Backx L, Schinzel A, Vermeesch JR, Zuffardi O. Cryptic deletions are a common finding in ‘balanced’ reciprocal and complex chromosome rearrangements: a study of 59 patients. J Med Genet 2007; 44: 750762.
  • 17
    Le Caignec C, Boceno M, Saugier-Veber P, Jacquemont S, Joubert M, David A, Frebourg T, Rival JM. Detection of genomic imbalances by array based comparative genomic hybridisation in fetuses with multiple malformations. J Med Genet 2005; 42: 121128.
  • 18
    Tyreman M, Abbott KM, Willatt LR, Nash R, Lees C, Whittaker J, Simonic I. High resolution array analysis: diagnosing pregnancies with abnormal ultrasound findings. J Med Genet 2009; 46: 531541.
  • 19
    Kleeman L, Bianchi DW, Shaffer LG, Rorem E, Cowan J, Craigo SD, Tighiouart H, Wilkins-Haug LE. Use of array comparative genomic hybridization for prenatal diagnosis of fetuses with sonographic anomalies and normal metaphase karyotype. Prenat Diagn 2009; 29: 12131217.
  • 20
    Vialard F, Molina GD, Leroy B, Quarello E, Escalona A, Le SC. Array comparative genomic hybridization in prenatal diagnosis: another experience. Fetal Diagn Ther 2009; 25: 277284.
  • 21
    Wellcome Trust Case Control Consortium. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 2007; 447: 661678.
  • 22
    Iafrate AJ, Feuk L, Rivera MN, Listewnik ML, Donahoe PK, Qi Y, Scherer SW, Lee C. Detection of large-scale variation in the human genome. Nat Genet 2004; 36: 949951.
  • 23
    Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, Moher D, Becker BJ, Sipe TA, Thacker SB. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000; 283: 20082012.
  • 24
    Redon R, Ishikawa S, Fitch KR, Feuk L, Perry GH, Andrews TD, Fiegler H, Shapero MH, Carson AR, Chen W, Cho EK, Dallaire S, Freeman JL, González JR, Gratacòs M, Huang J, Kalaitzopoulos D, Komura D, MacDonald JR, Marshall CR, Mei R, Montgomery L, Nishimura K, Okamura K, Shen F, Somerville MJ, Tchinda J, Valsesia A, Woodwark C, Yang F, Zhang J, Zerjal T, Zhang J, Armengol L, Conrad DF, Estivill X, Tyler-Smith C, Carter NP, Aburatani H, Lee C, Jones KW, Scherer SW, Hurles ME. Global variation in copy number in the human genome. Nature 2006; 444: 444454.
  • 25
    Kang TW, Jeon YJ, Jang E, Kim HJ, Kim JH, Park JL, Lee S, Kim YS, Kim JY, Kim SY. Copy number variations (CNVs) identified in Korean individuals. BMC Genomics 2008; 9: 492.
  • 26
    ACOG Commitee Opinion No. 446: array comparative genomic hybridization in prenatal diagnosis. Obstet Gynecol 2009; 114: 11611163.