[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Figure.  Flow Diagram of Study Selection
Flow Diagram of Study Selection
Table 1.  Characteristics of the Included Studies
Characteristics of the Included Studies
Table 2.  Summary of the Meta-analytic Pooled Proportions for Various Clinical Outcomes Among the Included Studies
Summary of the Meta-analytic Pooled Proportions for Various Clinical Outcomes Among the Included Studies
Table 3.  Types of OSD Reported Among the Included Studies
Types of OSD Reported Among the Included Studies
Table 4.  Pooled Incidence of Occult Spinal Dysraphism in Neonates and Infants With Various Cutaneous Stigmata According to Cutaneous Stigmataa
Pooled Incidence of Occult Spinal Dysraphism in Neonates and Infants With Various Cutaneous Stigmata According to Cutaneous Stigmataa
Supplement.

eAppendix. Supplementary Methods

eTable 1. Institutes Where Studies Were Conducted

eTable 2. Definitions of Tethered Cord, Low-Lying Conus Medullaris, and Thick Filum Terminale In Each Study and Data on Borderline Findings and Normal Variants on Screening Ultrasonography Excluded in Analysis of the Pooled Proportion of Definite Abnormality on Ultrasonography

eTable 3. Multiple Subgroup Analysis for Primary Outcomes

eFigure 1. Forest Plot and Funnel Plot of Pooled Incidence of Any Abnormal Ultrasonography Finding

eFigure 2. Forest Plot and Funnel Plot of Pooled Incidence of Definite Abnormal Finding on Ultrasonography

eFigure 3. Forest Plot and Funnel Plot of Pooled Incidence of Occult Spinal Dysraphism Among Total Patients

eFigure 4. Forest Plot and Funnel Plot of Pooled Proportion of Patients Managed With Surgical Intervention Among Patients Diagnosed With Occult Spinal Dysraphism

eFigure 5. Forest Plot and Funnel Plot of Pooled Proportion of Patients Managed With Surgical Intervention Among Total Patients

eFigure 6. Forest Plot of Pooled Incidence of Occult Spinal Dysraphism in Neonates and Infants with Single and Combined Stigmata

eFigure 7. Forest Plot of Pooled Incidence of Occult Spinal Dysraphism (OSD) in Neonates and Infants with Simple and Atypical Dimple

eFigure 8. Forest Plot of Pooled Incidence of Occult Spinal Dysraphism (OSD) and Pooled Proportions of Patients Managed With Neurosurgery Among Total Patients in Studies Including Low-Risk Cutaneous Stigmata and Intermediate-Risk Stigmata

eFigure 9. Forest Plot of Pooled Incidence of Occult Spinal Dysraphism (OSD) and Pooled Proportions of Patients Managed With Neurosurgery Among Total Patients Included in Studies that Performed MRI in All Cases With Abnormal Ultrasonography Results and Studies that Did Not Perform MRI in All Cases With Abnormal Ultrasonography Results

eReferences.

1.
Sewell  MJ, Chiu  YE, Drolet  BA.  Neural tube dysraphism: review of cutaneous markers and imaging.   Pediatr Dermatol. 2015;32(2):161-170. doi:10.1111/pde.12485PubMedGoogle Scholar
2.
Satar  N, Bauer  SB, Shefner  J, Kelly  MD, Darbey  MM.  The effects of delayed diagnosis and treatment in patients with an occult spinal dysraphism.   J Urol. 1995;154(2 Pt 2):754-758. doi:10.1016/S0022-5347(01)67154-4PubMedGoogle Scholar
3.
Gharedaghi  M, Samini  F, Mashhadinejad  H, Khajavi  M, Samini  M.  Orthopedic lesions in tethered cord syndrome: the importance of early diagnosis and treatment on patient outcome.   Arch Bone Jt Surg. 2014;2(2):93-97.PubMedGoogle Scholar
4.
Drolet  BA, Chamlin  SL, Garzon  MC,  et al.  Prospective study of spinal anomalies in children with infantile hemangiomas of the lumbosacral skin.   J Pediatr. 2010;157(5):789-794. doi:10.1016/j.jpeds.2010.07.054PubMedGoogle Scholar
5.
Allen  RM, Sandquist  MA, Piatt  JH  Jr, Selden  NR.  Ultrasonographic screening in infants with isolated spinal strawberry nevi.   J Neurosurg. 2003;98(3)(suppl):247-250. doi:10.3171/spi.2003.98.3.0247PubMedGoogle Scholar
6.
Ausili  E, Maresca  G, Massimi  L, Morgante  L, Romagnoli  C, Rendeli  C.  Occult spinal dysraphisms in newborns with skin markers: role of ultrasonography and magnetic resonance imaging.   Childs Nerv Syst. 2018;34(2):285-291. doi:10.1007/s00381-017-3638-0PubMedGoogle Scholar
7.
Ben-Amitai  D, Davidson  S, Schwartz  M,  et al.  Sacral nevus flammeus simplex: the role of imaging.   Pediatr Dermatol. 2000;17(6):469-471. doi:10.1046/j.1525-1470.2000.01824.xPubMedGoogle Scholar
8.
Ben-Sira  L, Ponger  P, Constantini  S.  Evaluation of dorsal midline discolorations with physical examination and ultrasound.   J Pediatr. 2017;190:246-250. doi:10.1016/j.jpeds.2017.07.023PubMedGoogle Scholar
9.
Ben-Sira  L, Ponger  P, Miller  E, Beni-Adani  L, Constantini  S.  Low-risk lumbar skin stigmata in infants: the role of ultrasound screening.   J Pediatr. 2009;155(6):864-869. doi:10.1016/j.jpeds.2009.06.003PubMedGoogle Scholar
10.
Chern  JJ, Kirkman  JL, Shannon  CN,  et al.  Use of lumbar ultrasonography to detect occult spinal dysraphism.   J Neurosurg Pediatr. 2012;9(3):274-279. doi:10.3171/2011.12.PEDS11351PubMedGoogle Scholar
11.
Choi  JH, Lee  T, Kwon  HH, You  SK, Kang  JW.  Outcome of ultrasonographic imaging in infants with sacral dimple.   Korean J Pediatr. 2018;61(6):194-199. doi:10.3345/kjp.2018.61.6.194PubMedGoogle Scholar
12.
Gibson  PJ, Britton  J, Hall  DM, Hill  CR.  Lumbosacral skin markers and identification of occult spinal dysraphism in neonates.   Acta Paediatr. 1995;84(2):208-209. doi:10.1111/j.1651-2227.1995.tb13612.xPubMedGoogle Scholar
13.
Henriques  JG, Pianetti  G, Henriques  KS, Costa  P, Gusmão  S.  Minor skin lesions as markers of occult spinal dysraphisms—prospective study.   Surg Neurol. 2005;63(suppl 1):S8-S12. doi:10.1016/j.surneu.2004.09.017PubMedGoogle Scholar
14.
Kriss  VM, Desai  NS.  Occult spinal dysraphism in neonates: assessment of high-risk cutaneous stigmata on sonography.   AJR Am J Roentgenol. 1998;171(6):1687-1692. doi:10.2214/ajr.171.6.9843314PubMedGoogle Scholar
15.
Kucera  JN, Coley  I, O’Hara  S, Kosnik  EJ, Coley  BD.  The simple sacral dimple: diagnostic yield of ultrasound in neonates.   Pediatr Radiol. 2015;45(2):211-216. doi:10.1007/s00247-014-3110-1PubMedGoogle Scholar
16.
McGovern  M, Mulligan  S, Carney  O, Wall  D, Moylett  E.  Ultrasound investigation of sacral dimples and other stigmata of spinal dysraphism.   Arch Dis Child. 2013;98(10):784-786. doi:10.1136/archdischild-2012-303564PubMedGoogle Scholar
17.
Robinson  AJ, Russell  S, Rimmer  S.  The value of ultrasonic examination of the lumbar spine in infants with specific reference to cutaneous markers of occult spinal dysraphism.   Clin Radiol. 2005;60(1):72-77. doi:10.1016/j.crad.2004.06.004PubMedGoogle Scholar
18.
Sneineh  AK, Gabos  PG, Keller  MS, Bowen  JR.  Ultrasonography of the spine in neonates and young infants with a sacral skin dimple.   J Pediatr Orthop. 2002;22(6):761-762. doi:10.1097/01241398-200211000-00013PubMedGoogle Scholar
19.
Wilson  P, Hayes  E, Barber  A, Lohr  J.  Screening for spinal dysraphisms in newborns with sacral dimples.   Clin Pediatr (Phila). 2016;55(11):1064-1070. doi:10.1177/0009922816664061PubMedGoogle Scholar
20.
Liberati  A, Altman  DG, Tetzlaff  J,  et al.  The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration.   BMJ. 2009;339:b2700. doi:10.1136/bmj.b2700PubMedGoogle Scholar
21.
IntHout  J, Ioannidis  JP, Borm  GF.  The Hartung-Knapp-Sidik-Jonkman method for random effects meta-analysis is straightforward and considerably outperforms the standard DerSimonian-Laird method.   BMC Med Res Methodol. 2014;14:25. doi:10.1186/1471-2288-14-25PubMedGoogle Scholar
22.
Higgins  JPT, Thompson  SG, Deeks  JJ, Altman  DG.  Measuring inconsistency in meta-analyses.   BMJ. 2003;327(7414):557-560. doi:10.1136/bmj.327.7414.557PubMedGoogle Scholar
23.
Egger  M, Davey Smith  G, Schneider  M, Minder  C.  Bias in meta-analysis detected by a simple, graphical test.   BMJ. 1997;315(7109):629-634. doi:10.1136/bmj.315.7109.629PubMedGoogle Scholar
24.
Medina  LS, Crone  K, Kuntz  KM.  Newborns with suspected occult spinal dysraphism: a cost-effectiveness analysis of diagnostic strategies.   Pediatrics. 2001;108(6):E101. doi:10.1542/peds.108.6.e101PubMedGoogle Scholar
25.
Guggisberg  D, Hadj-Rabia  S, Viney  C,  et al.  Skin markers of occult spinal dysraphism in children: a review of 54 cases.   Arch Dermatol. 2004;140(9):1109-1115. doi:10.1001/archderm.140.9.1109PubMedGoogle Scholar
26.
Sung  HJ, Lee  H-S.  Dorsal midline cutaneous stigmata associated with occult spinal dysraphism in pediatric patients.   Korean J Pediatr. 2019;62(2):68-74. doi:10.3345/kjp.2018.06744PubMedGoogle Scholar
27.
Sardana  K, Gupta  R, Garg  VK,  et al.  A prospective study of cutaneous manifestations of spinal dysraphism from India.   Pediatr Dermatol. 2009;26(6):688-695. doi:10.1111/j.1525-1470.2009.01014.xPubMedGoogle Scholar
28.
Zywicke  HA, Rozzelle  CJ.  Sacral dimples.   Pediatr Rev. 2011;32(3):109-113. doi:10.1542/pir.32-3-109PubMedGoogle Scholar
29.
Schropp  C, Sörensen  N, Collmann  H, Krauss  J.  Cutaneous lesions in occult spinal dysraphism—correlation with intraspinal findings.   Childs Nerv Syst. 2006;22(2):125-131. doi:10.1007/s00381-005-1150-4PubMedGoogle Scholar
30.
Lew  SM, Kothbauer  KF.  Tethered cord syndrome: an updated review.   Pediatr Neurosurg. 2007;43(3):236-248. doi:10.1159/000098836PubMedGoogle Scholar
31.
Chern  JJ, Aksut  B, Kirkman  JL,  et al.  The accuracy of abnormal lumbar sonography findings in detecting occult spinal dysraphism: a comparison with magnetic resonance imaging.   J Neurosurg Pediatr. 2012;10(2):150-153. doi:10.3171/2012.5.PEDS11564PubMedGoogle Scholar
32.
Cochrane Methods. Handbook for DTA reviews. Accessed May 28, 2020. https://methods.cochrane.org/sdt/handbook-dta-reviews
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Views 1,199
    Citations 0
    Original Investigation
    Pediatrics
    July 2, 2020

    Incidence of Occult Spinal Dysraphism Among Infants With Cutaneous Stigmata and Proportion Managed With Neurosurgery: A Systematic Review and Meta-analysis

    Author Affiliations
    • 1Research Institute of Radiology, Department of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea
    • 2Department of Radiology, Hallym University Medical Center, Dongtan Sacred Heart Hospital, Hwaseong, South Korea
    JAMA Netw Open. 2020;3(7):e207221. doi:10.1001/jamanetworkopen.2020.7221
    Key Points español 中文 (chinese)

    Question  What is the incidence of occult spinal dysraphism (OSD) in neonates or infants with various cutaneous stigmata, and how many of these cases are managed with a neurosurgical intervention?

    Findings  In this systematic review and meta-analysis of 15 studies involving 6558 patients, the pooled proportion of OSD among cases with cutaneous stigmata was 2.8%, and in 0.6% of patients, the condition was managed with neurological surgery. A stronger association with OSD was found in patients with combined stigmata and atypical dimple.

    Meaning  The findings of this study suggest that although the risk of OSD and the proportion of cases managed by neurosurgery were low, the characteristics of any cutaneous stigmata should be carefully evaluated in neonates or infants with midline cutaneous stigmata.

    Abstract

    Importance  Occult spinal dysraphism (OSD) is the most common congenital spinal anomaly. Cutaneous anomalies such as skin dimples or deviated gluteal folds are well known as stigmata of OSD and are indicators for further evaluation; however, the association between cutaneous anomalies and OSD has not been systemically evaluated.

    Objective  To evaluate the incidence of OSD and the proportion of OSD cases managed with a neurosurgical intervention among neonates or infants with various cutaneous stigmata.

    Data Sources  PubMed and Embase databases were searched for studies published up to July 25, 2018, that evaluated the proportion of OSD cases in neonates or infants with cutaneous stigmata. Search terms included ultrasound, dysraphism, dimple, and infant or neonate. The search was limited to English-language publications.

    Study Selection  Two reviewers selected the studies evaluating the incidence of OSD among neonates or infants with cutaneous stigmata.

    Data Extraction and Synthesis  The Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines for data extraction were followed. Pooled proportions of OSD cases and OSD cases that were managed with a neurosurgical intervention were obtained using the generalized linear mixed model and maximum likelihood method.

    Main Outcome and Measures  The pooled incidence of OSD and OSD cases managed with neurological surgery among patients with cutaneous stigmata was the primary outcome. This outcome was also evaluated in each subgroup, and heterogeneity was explored using subgroup analysis.

    Results  A total of 15 studies, involving 6558 neonate or infant patients with various cutaneous stigmata, were included. The pooled proportion of OSD cases among the patients with cutaneous stigmata was 2.8% (95% CI, 2.1%-3.8%; I2 = 51.6%), and the proportion managed with neurological surgery was 0.6% (95% CI, 0.3%-1.3%; I2 = 66.4%). Cases with combined stigmata showed a significantly higher association with OSD than those with a single stigma (10.5% [95% CI, 6.9%-15.8%] vs 2.3% [%, 95% CI, 1.5%-3.5%]; P < .001). The pooled proportion of OSD cases among patients with an atypical dimple was significantly higher than among those with simple dimple (8.8% [95% CI, 4.5%-16.6%] vs 0.6% [95% CI of 1.4%-2.1%]; P = .001).

    Conclusions and Relevance  The proportion of OSD in healthy, asymptomatic patients with midline cutaneous stigmata was low, and the proportion of patients who underwent a neurosurgical intervention was even lower. However, a careful evaluation as well as potential spinal magnetic resonance imaging is recommended for neonates or infants with combined stigmata or an atypical dimple for possible high-risk lesions.

    Introduction

    Occult spinal dysraphism (OSD) refers to a broad spectrum of skin-covered congenital spinal anomalies, including midline mesenchymal, neural, and bony elements. It is caused by an incomplete closure of the neural tube and anomalous development of the caudal cell mass during embryogenesis.1The clinical spectrum of OSD is broad, ranging from skin anomalies to motor, urinary, or bowel dysfunctions. Notably however, symptoms related to OSD are often not clinically obvious at birth and are usually subsequently revealed by a radiographic or physical examination. Therefore, affected patients present with delayed neurologic, urologic, and orthopedic symptoms and may have irreversible impairment.2,3

    Midline cutaneous lesions have long been recognized as stigmata of OSD and as indicators for conducting spinal ultrasonography in neonates. Approximately 50% to 80% of patients with OSD have cutaneous lesions.4 The identification of such cutaneous anomalies is crucial for the detection of OSD and enables early intervention and prevention of irreversible neurologic deterioration or infection.

    There have been multiple studies that evaluated the association between various kinds of midline cutaneous lesions and OSD.5-19 However, to our knowledge, the association between cutaneous anomalies and OSD has not been systemically evaluated. Therefore, the purpose of our current study was to evaluate the incidence of OSD in neonates and infants with various cutaneous stigmata as well as the proportion of these cases that were managed by neurosurgery.

    Methods
    Literature Search Strategy

    This study was conducted and written in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline.20 A computerized search of the PubMed and Embase databases was performed to identify studies that evaluated the frequency of OSD in neonates or infants with cutaneous stigmata. The cutoff date was July 25, 2018. The search terms used were ultraso* AND dysraphism OR dimple AND infant* OR neonate*. The bibliographies of these articles were also checked to identify other relevant articles. Our search was limited to publications in English.

    Inclusion and Exclusion Criteria

    Studies or subsets of studies that evaluated the proportion of OSD cases among neonates or infants with cutaneous stigmata were included if they matched all the following criteria: (1) neonates or infants with cutaneous lesions who underwent spinal sonography as an initial screening test for the evaluation of OSD; (2) studies that contained data on at least 10 consecutive patients; (3) observational studies; and (4) studies with data on the incidence of OSD. Studies were excluded if they met any of the following parameters: (1) a case report or case series with a small number of patients (ie, <10); (2) editorials, letters, comments, review articles, or conference abstracts; (3) studies providing insufficient data on the incidence of OSD; (4) studies of patients presenting with more than cutaneous lesions (eg, congenital anomalies); and (5) studies with overlapping data and patients. Two reviewers (H.M.Y., with 6 years experience in pediatric radiology, and S.J.C., with 3 years experience in radiology) independently evaluated the eligibility of the studies for inclusion.

    Data Extraction

    Data were extracted from the included studies using a standardized form. We compiled information on the following elements of study details: (1) study characteristics (authors, year of publication, duration of patient recruitment, affiliation, country of origin, study design), (2) demographic characteristics (sample size, male to female patient ratio, patient age at initial spinal sonography, and study population), and (3) use of spinal magnetic resonance imaging (MRI) as a reference standard. The following data on cutaneous stigmata was also extracted: (1) type of cutaneous stigmata, categorized into 3 groups (low risk [ie, simple dimple or deviated gluteal fold], intermediate risk [ie, vascular discoloration with or without low-risk stigmata], and high risk [ie, atypical dimple, hypertrichosis, pedunculated skin tag, fibroma pendulum, and any midline mass including lipoma]), and (2) number of cutaneous stigmata (ie, single vs combined). Finally, we extracted the following data on clinical outcomes: (1) number of patients who had any abnormal ultrasonography findings; (2) definition of tethered cord, low-lying conus medullaris level, and thick filum terminale; (3) number of patients with definite abnormal ultrasonography results after excluding any borderline findings (ie, borderline conus medullaris level or prominent filum terminale) or normal variations (ie, filar cyst); (4) number of patients diagnosed with OSD after excluding any borderline abnormal findings and normal variations; (5) individual types of OSD; and (6) number of patients managed with neurosurgery for OSD and their neurologic symptoms.

    Outcomes

    The primary outcomes of our current meta-analysis were the incidence of OSD among neonate or infant patients with cutaneous stigmata and the frequency of neurosurgical management among these cases. The secondary outcomes were the results of subgroup analyses for the primary outcomes according to the number (ie, single vs combined) and type (ie, simple dimple vs atypical dimple and 3 risk groups) of stigmata and whether a reference standard (ie, spinal MRI) was used for every patient with abnormal findings on spinal ultrasonography or for a proportion of patients with abnormal ultrasonography findings.

    Statistical Analysis

    The pooled proportions obtained from the included studies were analyzed using the generalized linear mixed model and maximum likelihood method. We used the Hartung-Knapp-Sidik-Jonkman method to derive confidence interval.21 Heterogeneity was assessed using the inconsistency index (I2) established by Higgins et al,22 which reflects the percentage of variation across studies that is caused by heterogeneity rather than chance. The values of more than 50% indicating the presence of substantial heterogeneity. To address possible publication bias, a funnel plot was visually evaluated and the Egger test which is a test for asymmetry of the funnel plot was used for the statistical assessment.23 The meta package in R version 3.6.2 (R Project for Statistical Computing) was used. Statistical significance was set at P <.05, and all tests were 2-tailed.

    Results
    Characteristics of the Included Studies

    The process of literature search is summarized in the eAppendix in the Supplement, and the Figure illustrates the selection process. The characteristics of the 15 included studies are listed in Table 1, and the institutions at which the studies were conducted are detailed in eTable 1 in the Supplement. Among these publications, 6 studies were prospective6-9,13,14 and 9 were retrospective.5,10-12,15-19 The number of patients who underwent spinal ultrasonography screening ranged from 16 to 3884. The reported mean age of the patients at the time of ultrasonography examination ranged from 1.77 to 52.8 days but was not available for 8 studies.5,7,12-17 The main indications for an ultrasonogram evaluation were a low-risk stigmata in 3 studies,5,7,8 an intermediate-risk stigmata in 3 studies9,15,18 and low- to high-risk stigmata in 9 studies.6,10-14,16,17,19 Regarding a reference standard for the imaging diagnosis of OSD, the proportion of patients undergoing spinal MRI ranged from 0.4% to 31.3%.5-13,16,17,19 Spinal MRI was not performed in 1 study.18 All patients with an abnormal finding on ultrasonography underwent spinal MRI in 7 studies,6-10,13,14 while with a portion of patients with an abnormal finding on ultrasonography underwent MRI in 6 studies.11,12,15-17,19

    Meta-analytic Pooled Incidence of OSD and Pooled Proportion of Patients Managed With Neurosurgery

    The pooled clinical outcomes for the 15 included studies are summarized in Table 2. A total of 6558 patients examined in the included study series underwent spinal ultrasonography, among whom 426 (6.5%) had abnormal sonographic results. These abnormal results included borderline findings and normal variations, such as borderline conus medullaris level, prominent filum terminale, or filar cyst.6,8-11 The pooled incidence of any abnormal finding, including borderline findings and normal variations, was 5.3% (95% CI, 2.6%-10.5%). Definitions of low-lying conus medullaris and thickness of filum terminale as well as detailed information regarding excluded data of borderline results and normal variants are summarized in eTable 2 in the Supplement. The commonly used criteria for the level of low-lying conus medullaris and thickness of filum terminale were below the L2-3 intervertebral disc space and 2 mm, respectively. The definition of tethered cord was missing or vague in 11 of 15 studies (eTable 2 in the Supplement). After excluding these borderline results and normal variants, a total of 229 of 5615 patients showed a definite abnormal result on spinal ultrasonography. The pooled incidence of sonographic abnormalities was 3.1% (95% CI, 1.6%-6.0%). In 14 of 15 studies, 164 patients were diagnosed with OSD by ultrasonography and/or follow-up MRI, and the overall pooled incidence was 2.8% (95% CI, 2.4%-3.8%; I2 = 51.6%). Three studies did not report on the specific types of OSD,10,14,19 and the details of 128 OSD cases in the 11 other studies are described in Table 3. We found that 26 of 6364 patients underwent neurologic surgery, a pooled proportion of 0.6% (95% CI, 0.3%-1.3%; I2 = 66.4%). The relevant forest plots and funnel plots are shown in eFigures 1-5 in the Supplement. The preoperative symptoms of these cases were described in only 3 studies,7,15,19 and these neurologic symptoms included unremarkable (5 patients) and intermittent urinary and fecal incontinence (1 patient).

    Compared with the other studies (range, 0%-5.3%), 1 study reported a relatively high incidence of OSD (2 of 16 patients [12.5%]).5 The study was performed in patients with strawberry nevus.

    Multiple Subgroup Analysis

    In a head-to-head comparison of 7 studies,5,6,8,13,16-18 the incidence of OSD was significantly higher among neonates and infants with combined stigmata (10.5%; 95% CI, 6.9%-15.8%, I2 = 38.4%) than among those with single stigmata (2.3%, 95% CI, 1.5%-3.5%, I2 = 18.8%) (P < .001) (Table 4). Compared with patients with an atypical dimple, those with a simple dimple showed a relatively low rate of OSD (8.8% [95% CI, 4.5%-16.6%] vs 0.6% [95% CI of 1.4%-2.1%]; P = .001).

    Regarding the risk of cutaneous stigmata, 3 studies examined cases of low-risk stigmata,9,15,18 and a further 3 reports included patients with intermediate-risk stigmata, with or without a low-risk lesion.5,7,8 There was no study that included only patients with high-risk stigmata. There was no statistically significant difference between the pooled OSD incidence in studies of intermediate-risk stigmata cases (2.8%; 95% CI, 1.1%-7.3%) compared with those of low-risk stigmata cases (0.6%; 95% CI, 0.2%-14.5%) (P = .36). The pooled rate of neurosurgery was significantly higher in studies that included intermediate-risk stigmata cases (2.1%; 95% CI, 0.2%-16.3%) compared with those that included low-risk stigmata cases (0.1%; 95% CI, 0.1%-0.3%) (P = .02) (eTable 2 in the Supplement).

    We classified our studies in 2 groups according to whether all patients with abnormal sonographic findings underwent a spinal MRI. There was no statistically significant difference between the incidence of OSD in the 6 studies in which all patients with abnormal sonographic findings underwent MRI (2.7%; 95% CI, 1.4%-5.2%) compared with that in 6 other studies in which spinal MRI had not been routinely used (2.2%; 95% CI, 1.8%-2.7%) (P = .96). There was also no statistically significant difference between the pooled proportion of patients managed with neurosurgery in the former subgroup of studies (0.8%; 95% CI, 0.5%-1.4%) compared with the latter subgroup (0.2%; 95% CI, 0.1%-0.4%) (P = .08) (eTable 2 in the Supplement). All relevant forest plots are shown in eFigures 6-9 in the Supplement.

    Discussion

    Spinal ultrasonography is generally preferred as a first-line imaging test for low-risk children with suspected OSD because of its cost-effectiveness,24 accessibility, no requirement for sedation, and good accuracy through normally incompletely ossified posterior elements.9 The wide use of spinal imaging in infants with cutaneous lesions requires an investigation of the evidence base for the prevalence of OSD in this population. In addition, the aggregate data from the 15 studies, involving 6558 patients, in our current systematic review and meta-analysis have revealed a 3.1% pooled proportion of OSD among patients with definite abnormal results on spinal ultrasonography and a pooled OSD rate as low as 2.8%. In addition, few of the neonates or infants examined in these prior reports underwent a neurosurgical intervention, with a pooled proportion of 0.6%.

    The prevalence of OSD was significantly higher in patients with combined stigmata (10.5%) than those with a single stigmata (2.3%). This finding is consistent with the results of previous studies that reported that a combination of 2 or more midline cutaneous lesions is more likely to be associated with OSD than a single lesion.25,26 Therefore, when evaluating infants with dorsal midline skin lesions, meticulous screening for skin lesions is warranted to determine whether 2 or more lesions are present, given that this indicates a higher risk of OSD. However, some studies in our current review focused on specific skin lesions, such as spinal strawberry nevi5 and dorsal midline vascular anomalies,8 that are associated with other stigmata. In 1 study performed among patients with strawberry nevus,5 the incidence of OSD was higher (12.5%) than that reported in other studies. Therefore, there is a possibility that the types of skin lesions, rather than simply a combination of multiple lesions, could affect OSD outcomes. A future large-scale study to further elucidate the clinical significance of the different types of specific stigmata and their associations with OSD is warranted.

    The pooled incidence of OSD in the patients with an atypical dimple in our included studies was significantly higher than that in patients with a simple dimple (8.8% vs 0.6%). This result was in accordance with that of a previous study,27 which reported atypical dimple as a high-risk factor for OSD. It is generally accepted that further evaluations via radiologic imaging and early neurosurgical referrals are required when atypical dimples exist.28 The most commonly used criteria for defining simple dimples are a small size (ie, <5 mm) with a midline placement within 2.5 cm of the anus and no association of other cutaneous stigmata.6,14,25 An atypical dimple is larger than 5 mm and located within 25 mm of the anus.4,17 Other criteria for an atypical dimple include deep dimple,12,13 dimples located cranially to the gluteal crease or outside the midline,29 and multiple dimples.28 Therefore, clinicians involved in the referral and management of infants with a sacral dimple need to be well informed about the characteristics of atypical dimples, and careful examinations should be performed to ascertain the atypical features of such lesions in clinical practice.

    In terms of the degree of OSD risk for patients with cutaneous stigmata, there was no statistically significant difference between the intermediate-risk group and the low-risk group. However, the pooled proportion of infants and neonates who underwent neurosurgery was significantly higher among patients with intermediate-risk stigmata than among those with low-risk stigmata (2.1% vs 0.1%). The risk stratifications based on lumbosacral midline cutaneous lesions might better reflect the presence of clinically significant OSD requiring surgery than a mere diagnosis of OSD. However, care must still be taken with these interpretations because there have been only a small number of studies for each risk group to date.

    In our current meta-analysis, various types of OSD were shown to be associated with cutaneous stigmata. The most common type was low-lying conus medullaris (or tethered cord), followed by pathologic filum terminale and dermal sinus. Therefore, in the imaging evaluation of patients with suspected OSD, the level of the conus medullaris, motion of the conus medullaris and nerve roots, morphology of the filum terminale, and tract of dorsal dermal sinus should be carefully assessed. In the included studies, the commonly used criteria for the level of low-lying conus medullaris and thickness of filum terminale were below the L2-3 intervertebral disc space and 2 mm, respectively. Regarding the definition of tethered cord, there appeared to be no consensus among studies, and its definition was missing or vague in most studies included in this meta-analysis (11 of 15). We accepted whatever definition of low-lying conus medullaris or tethered cord provided by the authors of the included studies to determine the pooled proportion of OSD and its subtypes. This might limit our results.

    We aimed to investigate the number of patients with OSD who underwent neurosurgical management and to identify the main cause of surgical treatment. Only 0.6% of patients who had cutaneous stigmata underwent neurosurgical intervention, and there was limited information about preoperative symptoms. Preoperative symptoms were described in only 6 patients, and only 1 had clinically noticeable symptoms. These results did not mean that this proportion of patients truly represented the number of patients who required neurosurgical management. There was no clear indication of surgical intervention in asymptomatic patients with apparent tethered cord as observed on imaging30; thus, the decision to perform surgical treatment was based on the subjective preference of surgeons and patients’ guardians. The true prevalence of neurologic impairment in patients with cutaneous stigmata who require surgical intervention should be further investigated.

    When OSD is suspected on a spinal ultrasonography, a spinal MRI is regarded as the criterion-standard modality for further clarification and surgical planning. Previous studies have reported a lower sensitivity of ultrasonography compared with MRI for detecting various types of OSD, such as dermal sinus tract, thickened filum, or fatty filum.9,31 However, because of high costs, relatively poor availability, and the need for sedation, MRI has a more limited application in some cases. Considering this situation, we divided included studies into 2 groups, stratified by whether spinal MRI was done for every case with an abnormal ultrasonography findings. Studies in which a spinal MRI was conducted for every patient with an abnormal spinal ultrasonography finding did not have a statistically significantly higher pooled proportion of cases with a diagnosis of OSD (2.7% vs 2.2%) or cases managed with neurosurgery (0.8% vs 0.2%). This finding suggests that a spinal MRI should be given careful consideration when there are increased-risk conditions and when ultrasonography results are abnormal for an early and accurate detection of OSD and to minimize the possible neurologic impairment in affected patients.

    Limitations

    This study has several potential limitations of note. First, most included studies were retrospective (9 of 15). However, most reports included a relatively large number of patients (ranging from 94 to 3884) with only 2 studies5,18 investigating small cohorts (16 and 50 cases). Second, a considerable heterogeneity was noted among the included studies. Thus, we performed multiple subgroup analyses. Third, we did not conduct a quality assessment of individual studies. Because of the observational nature of the included studies, there was no appropriate evaluation tool for quality assessment. To overcome these issues, we used a currently available robust methodology and followed the PRISMA guideline20 and the Cochrane Collaboration 2013 Handbook for Diagnostic Test Accuracy Reviews32 for this study.

    Conclusions

    In this study, the prevalence of OSD among otherwise healthy infants and newborns with midline cutaneous stigmata was as low as 2.8%, and the cases that were managed with a neurosurgical intervention was even lower, at less than 1%, in a spinal ultrasonography screening setting. Pediatricians, neurosurgeons, dermatologists, radiologists, and other clinicians involved in the management of children with dorsal midline cutaneous stigmata should carefully evaluate whether these patients have higher-risk stigmata and then consider a spinal MRI in such cases to screen for possible OSD.

    Back to top
    Article Information

    Accepted for Publication: March 25, 2020.

    Published: July 2, 2020. doi:10.1001/jamanetworkopen.2020.7221

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Choi SJ et al. JAMA Network Open.

    Corresponding Author: Hee Mang Yoon, MD, PhD, Research Institute of Radiology, Department of Radiology, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea (espoirhm@gmail.com).

    Author Contributions: Dr Yoon had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Yoon, Suh, Cho, Lee.

    Acquisition, analysis, or interpretation of data: Choi, Yoon, Hwang, Suh, Jung, Lee.

    Drafting of the manuscript: Choi, Yoon, Hwang, Suh, Cho, Lee.

    Critical revision of the manuscript for important intellectual content: Yoon, Suh, Jung, Lee.

    Statistical analysis: Suh.

    Obtained funding: Yoon, Cho.

    Administrative, technical, or material support: Choi, Cho, Lee.

    Supervision: Yoon, Hwang, Suh, Jung, Lee.

    Conflict of Interest Disclosures: None reported.

    Funding/Support: This study was supported by grant 2020IE0016 from Asan Medical Center Children’s Hospital (Heart Institute) and grant 2020R1G1A1004591 from the National Research Foundation of Korea.

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    References
    1.
    Sewell  MJ, Chiu  YE, Drolet  BA.  Neural tube dysraphism: review of cutaneous markers and imaging.   Pediatr Dermatol. 2015;32(2):161-170. doi:10.1111/pde.12485PubMedGoogle Scholar
    2.
    Satar  N, Bauer  SB, Shefner  J, Kelly  MD, Darbey  MM.  The effects of delayed diagnosis and treatment in patients with an occult spinal dysraphism.   J Urol. 1995;154(2 Pt 2):754-758. doi:10.1016/S0022-5347(01)67154-4PubMedGoogle Scholar
    3.
    Gharedaghi  M, Samini  F, Mashhadinejad  H, Khajavi  M, Samini  M.  Orthopedic lesions in tethered cord syndrome: the importance of early diagnosis and treatment on patient outcome.   Arch Bone Jt Surg. 2014;2(2):93-97.PubMedGoogle Scholar
    4.
    Drolet  BA, Chamlin  SL, Garzon  MC,  et al.  Prospective study of spinal anomalies in children with infantile hemangiomas of the lumbosacral skin.   J Pediatr. 2010;157(5):789-794. doi:10.1016/j.jpeds.2010.07.054PubMedGoogle Scholar
    5.
    Allen  RM, Sandquist  MA, Piatt  JH  Jr, Selden  NR.  Ultrasonographic screening in infants with isolated spinal strawberry nevi.   J Neurosurg. 2003;98(3)(suppl):247-250. doi:10.3171/spi.2003.98.3.0247PubMedGoogle Scholar
    6.
    Ausili  E, Maresca  G, Massimi  L, Morgante  L, Romagnoli  C, Rendeli  C.  Occult spinal dysraphisms in newborns with skin markers: role of ultrasonography and magnetic resonance imaging.   Childs Nerv Syst. 2018;34(2):285-291. doi:10.1007/s00381-017-3638-0PubMedGoogle Scholar
    7.
    Ben-Amitai  D, Davidson  S, Schwartz  M,  et al.  Sacral nevus flammeus simplex: the role of imaging.   Pediatr Dermatol. 2000;17(6):469-471. doi:10.1046/j.1525-1470.2000.01824.xPubMedGoogle Scholar
    8.
    Ben-Sira  L, Ponger  P, Constantini  S.  Evaluation of dorsal midline discolorations with physical examination and ultrasound.   J Pediatr. 2017;190:246-250. doi:10.1016/j.jpeds.2017.07.023PubMedGoogle Scholar
    9.
    Ben-Sira  L, Ponger  P, Miller  E, Beni-Adani  L, Constantini  S.  Low-risk lumbar skin stigmata in infants: the role of ultrasound screening.   J Pediatr. 2009;155(6):864-869. doi:10.1016/j.jpeds.2009.06.003PubMedGoogle Scholar
    10.
    Chern  JJ, Kirkman  JL, Shannon  CN,  et al.  Use of lumbar ultrasonography to detect occult spinal dysraphism.   J Neurosurg Pediatr. 2012;9(3):274-279. doi:10.3171/2011.12.PEDS11351PubMedGoogle Scholar
    11.
    Choi  JH, Lee  T, Kwon  HH, You  SK, Kang  JW.  Outcome of ultrasonographic imaging in infants with sacral dimple.   Korean J Pediatr. 2018;61(6):194-199. doi:10.3345/kjp.2018.61.6.194PubMedGoogle Scholar
    12.
    Gibson  PJ, Britton  J, Hall  DM, Hill  CR.  Lumbosacral skin markers and identification of occult spinal dysraphism in neonates.   Acta Paediatr. 1995;84(2):208-209. doi:10.1111/j.1651-2227.1995.tb13612.xPubMedGoogle Scholar
    13.
    Henriques  JG, Pianetti  G, Henriques  KS, Costa  P, Gusmão  S.  Minor skin lesions as markers of occult spinal dysraphisms—prospective study.   Surg Neurol. 2005;63(suppl 1):S8-S12. doi:10.1016/j.surneu.2004.09.017PubMedGoogle Scholar
    14.
    Kriss  VM, Desai  NS.  Occult spinal dysraphism in neonates: assessment of high-risk cutaneous stigmata on sonography.   AJR Am J Roentgenol. 1998;171(6):1687-1692. doi:10.2214/ajr.171.6.9843314PubMedGoogle Scholar
    15.
    Kucera  JN, Coley  I, O’Hara  S, Kosnik  EJ, Coley  BD.  The simple sacral dimple: diagnostic yield of ultrasound in neonates.   Pediatr Radiol. 2015;45(2):211-216. doi:10.1007/s00247-014-3110-1PubMedGoogle Scholar
    16.
    McGovern  M, Mulligan  S, Carney  O, Wall  D, Moylett  E.  Ultrasound investigation of sacral dimples and other stigmata of spinal dysraphism.   Arch Dis Child. 2013;98(10):784-786. doi:10.1136/archdischild-2012-303564PubMedGoogle Scholar
    17.
    Robinson  AJ, Russell  S, Rimmer  S.  The value of ultrasonic examination of the lumbar spine in infants with specific reference to cutaneous markers of occult spinal dysraphism.   Clin Radiol. 2005;60(1):72-77. doi:10.1016/j.crad.2004.06.004PubMedGoogle Scholar
    18.
    Sneineh  AK, Gabos  PG, Keller  MS, Bowen  JR.  Ultrasonography of the spine in neonates and young infants with a sacral skin dimple.   J Pediatr Orthop. 2002;22(6):761-762. doi:10.1097/01241398-200211000-00013PubMedGoogle Scholar
    19.
    Wilson  P, Hayes  E, Barber  A, Lohr  J.  Screening for spinal dysraphisms in newborns with sacral dimples.   Clin Pediatr (Phila). 2016;55(11):1064-1070. doi:10.1177/0009922816664061PubMedGoogle Scholar
    20.
    Liberati  A, Altman  DG, Tetzlaff  J,  et al.  The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration.   BMJ. 2009;339:b2700. doi:10.1136/bmj.b2700PubMedGoogle Scholar
    21.
    IntHout  J, Ioannidis  JP, Borm  GF.  The Hartung-Knapp-Sidik-Jonkman method for random effects meta-analysis is straightforward and considerably outperforms the standard DerSimonian-Laird method.   BMC Med Res Methodol. 2014;14:25. doi:10.1186/1471-2288-14-25PubMedGoogle Scholar
    22.
    Higgins  JPT, Thompson  SG, Deeks  JJ, Altman  DG.  Measuring inconsistency in meta-analyses.   BMJ. 2003;327(7414):557-560. doi:10.1136/bmj.327.7414.557PubMedGoogle Scholar
    23.
    Egger  M, Davey Smith  G, Schneider  M, Minder  C.  Bias in meta-analysis detected by a simple, graphical test.   BMJ. 1997;315(7109):629-634. doi:10.1136/bmj.315.7109.629PubMedGoogle Scholar
    24.
    Medina  LS, Crone  K, Kuntz  KM.  Newborns with suspected occult spinal dysraphism: a cost-effectiveness analysis of diagnostic strategies.   Pediatrics. 2001;108(6):E101. doi:10.1542/peds.108.6.e101PubMedGoogle Scholar
    25.
    Guggisberg  D, Hadj-Rabia  S, Viney  C,  et al.  Skin markers of occult spinal dysraphism in children: a review of 54 cases.   Arch Dermatol. 2004;140(9):1109-1115. doi:10.1001/archderm.140.9.1109PubMedGoogle Scholar
    26.
    Sung  HJ, Lee  H-S.  Dorsal midline cutaneous stigmata associated with occult spinal dysraphism in pediatric patients.   Korean J Pediatr. 2019;62(2):68-74. doi:10.3345/kjp.2018.06744PubMedGoogle Scholar
    27.
    Sardana  K, Gupta  R, Garg  VK,  et al.  A prospective study of cutaneous manifestations of spinal dysraphism from India.   Pediatr Dermatol. 2009;26(6):688-695. doi:10.1111/j.1525-1470.2009.01014.xPubMedGoogle Scholar
    28.
    Zywicke  HA, Rozzelle  CJ.  Sacral dimples.   Pediatr Rev. 2011;32(3):109-113. doi:10.1542/pir.32-3-109PubMedGoogle Scholar
    29.
    Schropp  C, Sörensen  N, Collmann  H, Krauss  J.  Cutaneous lesions in occult spinal dysraphism—correlation with intraspinal findings.   Childs Nerv Syst. 2006;22(2):125-131. doi:10.1007/s00381-005-1150-4PubMedGoogle Scholar
    30.
    Lew  SM, Kothbauer  KF.  Tethered cord syndrome: an updated review.   Pediatr Neurosurg. 2007;43(3):236-248. doi:10.1159/000098836PubMedGoogle Scholar
    31.
    Chern  JJ, Aksut  B, Kirkman  JL,  et al.  The accuracy of abnormal lumbar sonography findings in detecting occult spinal dysraphism: a comparison with magnetic resonance imaging.   J Neurosurg Pediatr. 2012;10(2):150-153. doi:10.3171/2012.5.PEDS11564PubMedGoogle Scholar
    32.
    Cochrane Methods. Handbook for DTA reviews. Accessed May 28, 2020. https://methods.cochrane.org/sdt/handbook-dta-reviews
    ×