These are the two article for today’s journal club i am supposed to hive comment as an audience for each study separately. Tell me what are the setbacks of these studies individually other than those already mentioned in the article… and also tell how they could have avoided or improved those setbacks
Reading File
Reading File
Browsing Directory
~/
Searching the Web
Segmental infantile hemangiomas midline LUMBAR syndrome Metry 2026 Pediatric Dermatology
Searching PubMed
infantile hemangioma LUMBAR
Searching PubMed
pediatric vulvar lichen sclerosus
Browsing Directory
~/attachments/3d266c24-72ec-4db1-bdd5-f692a7aac34a/
Reading Web Page
https://pubmed.ncbi.nlm.nih.gov/42021702
Finding Sources
Finding Sources
Running Command
python -c "from pathlib import Path; from pypdf import PdfReader; base=Path('/tmp/workspace/attachments/3d266c24-72ec-4db1-bdd5-f692a7aac34a'); out=Path('/tmp/workspace/journal-club'); out.mkdir(exist_ok=True); files=list(base.glob('*.pdf')); [(out/(f.stem+'.txt')).write_text('\\n\\n'.join('--- PAGE %d ---\\n%s'%(i+1,(p.extract_text() or '')) for i,p in enumerate(PdfReader(str(f)).pages)),encoding='utf8') for f in files]; print('\\n'.join(str(p) for p in out.iterdir()))"
Reading File
Reading File
Searching the Web
"Long-Term Outcomes of Prepubertal-Onset Vulvar Lichen Sclerosus" Savage
Searching Content
"limitation" in ~/journal-club/
Searching Content
"Methods" in ~/journal-club/
Searching Content
"Discussion" in ~/journal-club/
Searching Content
"Table 1" in ~/journal-club/
Reading Article
PMID: 41084133
Reading Article
PMID: 42021702
Reading File
Reading File
Reading File
Reading File
Reading File
Reading File
Reading File
1. Metry et al. - Segmental infantile hemangiomas involving the midline and LUMBAR syndrome
The authors already acknowledge publication/selection bias and incomplete photographic availability (91 of 148 identified LUMBAR cases) as limitations (p. 897). Additional audience-level critiques:
| Additional setback | Why it matters | How it could be avoided or improved |
|---|---|---|
| No comparison group of infants with segmental but strictly lateral hemangiomas who underwent systematic LUMBAR screening | The study demonstrates that midline involvement was present among known LUMBAR cases. It does not directly quantify the risk of LUMBAR in a child with a lateral lesion, nor prove that the risk is negligible. “No cases” in this selected series is not the same as a validated rule-out criterion. | Conduct a prospective multicenter cohort enrolling consecutive infants with all lumbosacral/pelvic segmental IH patterns, irrespective of location, with uniform spinal, renal, pelvic and urogenital assessment. Calculate sensitivity, specificity, predictive values, and confidence intervals for midline involvement. |
| Case-only design cannot establish predictive performance | A diagnostic-risk claim needs both affected and unaffected children. This study can support a phenotype associated with LUMBAR, but cannot determine whether midline involvement independently predicts the syndrome. | Use a diagnostic-accuracy or cohort design including children with and without LUMBAR. A multivariable model could assess morphology, size, subsite, ulceration, IH-MAG phenotype, sex, and midline involvement together. |
| Potential incorporation or circularity bias | Cases were obtained partly from the dataset used to establish LUMBAR diagnostic criteria. Since a segmental lower-body IH is part of the syndrome definition, finding that all LUMBAR cases had segmental IH is partly built into how cases are classified. This makes the conclusion about “segmental” morphology less independent. | Validate the proposed midline criterion in an external cohort, ideally assessed before a final LUMBAR diagnosis is assigned. Use blinded adjudication by clinicians not involved in defining the diagnostic criteria. |
| Photo-based classification is subjective and lacks formal reliability data | “Segmental,” “partial segmental,” “localized,” “transmedian,” and “paramedian” can be difficult to distinguish, particularly on the lower body where maps are not established. Four uncertain cases underwent author consensus, but there is no kappa statistic or formal inter-rater agreement for the whole sample. | Predefine photographic classification criteria, use standardized body maps, have multiple independent blinded assessors classify every image, and report inter-rater agreement. Include measurements or image overlays showing the anatomical midline. |
| Assessors may have been aware that every child had LUMBAR syndrome | Knowledge of the diagnosis can unconsciously influence morphology classification, especially in equivocal images. | Blind image assessors to syndrome status and clinical/anomaly data. Mix photographs from children with LUMBAR, isolated IH, and other vascular lesions. |
| Clinical phenotype may have been incompletely captured by a single or late photograph | Infantile hemangiomas change substantially during proliferation and involution. A lesion classified from one photograph could appear more or less segmental at another age. The study does not report a standardized age at photography. | Prospectively obtain serial standardized photographs at presentation and during proliferation, with the child positioned consistently and the entire relevant anatomical territory visible. |
| Published cases may not represent ordinary clinical practice | Published LUMBAR cases may be unusually severe, atypical, or memorable, with more extensive anomalies and more striking hemangiomas. This may inflate the apparent uniformity of the phenotype. | Establish a population-based or registry-based prospective study that captures consecutive referrals and community cases, including less severe and imaging-negative cases. |
| The “strictly lateral lesions do not warrant evaluation” conclusion is stronger than the data support | The data are consistent with low risk, but the sample is only 91 photo-confirmed LUMBAR cases. A rare exception could easily be missed. It also should not override evaluation prompted by neurological, urogenital, anorectal, renal, or bony findings. | Phrase the clinical implication more cautiously: “isolated strictly lateral lesions appear lower risk in this dataset.” Develop an algorithm that retains imaging for children with concerning associated anomalies, uncertain morphology, or inadequate clinical examination. |
A concise journal-club comment
This study identifies a convincing common phenotype among children already diagnosed with LUMBAR syndrome, but it does not directly validate midline involvement as a screening test. Because it lacks a systematically screened non-LUMBAR comparison group, the recommendation to forgo evaluation in strictly lateral lesions should be considered hypothesis-generating rather than a definitive rule-out criterion.
2. Savage et al. - Long-term outcomes of prepubertal-onset vulvar lichen sclerosus
The authors acknowledge small sample size, single-center recruitment, variable follow-up duration, and failure of some patients to return for evaluation (p. 282). Additional issues:
| Additional setback | Why it matters | How it could be avoided or improved |
|---|---|---|
| Strong selection and attrition bias from inclusion of only 68 of 135 screened records | Only about half of screened patients entered the final analysis. Participants who could be contacted, consented, completed questionnaires, and attended review may differ materially from nonparticipants in disease severity, adherence, access to care, language, health literacy, or current symptoms. The reported 62% remission rate may therefore not represent the original clinic cohort. | Report baseline characteristics and outcomes, where available, for included versus excluded/nonresponding patients. Use a prospective inception cohort with active follow-up procedures and linkage to records to reduce loss to follow-up. |
| Exclusion of non-English-speaking patients limits equity and generalizability | Language is linked to health literacy, access, social context, and treatment adherence. Excluding these patients may systematically select a more resourced group and underrepresent families with barriers to care. | Use professionally translated questionnaires, interpreters, and culturally adapted patient materials. Record language and socioeconomic indicators as potential determinants of adherence and outcome. |
| Adherence is broadly defined and likely measured by recall/self-report | “Fully adherent,” “partially adherent,” and “non-adherent” categories are vulnerable to recall error and social-desirability bias. The category “partially adherent” is especially broad, potentially combining clinically very different levels of treatment exposure. | Prospectively record adherence at scheduled intervals using a structured diary, electronic reminders/monitoring, prescription refill data, medication-weight measurement, and a validated adherence instrument. Analyze adherence as a quantitative, time-varying exposure where feasible. |
| Association between adherence and remission cannot establish causation | Patients with milder disease may find treatment easier to continue and may be more likely to remit anyway. Conversely, patients with more severe, painful, or treatment-resistant disease may appear “non-adherent” because they need more treatment, discontinue because of poor response, or have more follow-up. This is confounding by disease severity and reverse causation. | Measure and adjust for baseline clinical severity, anatomic involvement, symptoms, time to diagnosis, family factors, treatment adverse effects, and socioeconomic access. A prospective study with repeated severity and adherence measures would better establish temporal sequence. |
| The fully adherent group is very small: 13 patients | The striking remission figure of 92.3% is based on 12 of 13 patients. Such a result has a wide confidence interval and is unstable: one or two different outcomes would change the estimate considerably. | Present confidence intervals around all proportions and effect estimates. Pool data across centers, preregister analyses, and ensure adequate sample size before fitting regression models. |
| Potential overfitting and underpowered regression analyses | The cohort has relatively few outcome events and important missing data. Logistic regression with several candidate predictors can generate unstable adjusted associations, especially when the fully adherent category contains only 13 patients. | Limit predictors according to an a priori statistical plan, report events-per-parameter and confidence intervals, use penalized regression when appropriate, and validate models externally. |
| Outcome ascertainment was not uniform | Persistence/recurrence could be identified from questionnaire and/or clinician examination, whereas VQLI was completed only by adults and PGA comparison required available photographs. Different assessment pathways can lead to differential misclassification. | Use the same scheduled clinical examination and validated patient-reported outcomes for every participant. Prospectively collect standardized baseline and follow-up photographs and use blinded outcome adjudication. |
| “Remission” may be misclassified because disease activity can be intermittent | Remission was defined as no symptoms/signs and no treatment at assessment. In a relapsing condition, a symptom-free interval does not necessarily mean durable remission. Variable intervals between visits make this concern larger. | Define sustained remission prospectively, for example absence of symptoms and objective signs without treatment for a prespecified period, and use survival analysis for time to remission and time to relapse. |
| Menarche was imputed as age 12 when not documented | Menarche is central to the study question. Assigning an arbitrary age can misclassify pre- versus post-menarche timing and alter estimates of disease persistence around puberty. Although those with imputed values were excluded from one analysis, this does not remove the concern from all analyses. | Retrieve menarche timing through direct patient/parent confirmation or linked primary-care records. If data remain missing, use multiple imputation and sensitivity analyses rather than a fixed assumed age. |
| Treatment exposure was not standardized over the 2003-2022 period | The practice protocol is described, but actual potency, dose, maintenance regimen, duration, counseling, treatment changes, and co-interventions may have differed over almost two decades. “Adherence” to a changing regimen is not a uniform exposure. | Prospectively standardize first-line and maintenance protocols, document cumulative topical corticosteroid exposure and treatment modifications, and analyze regimen intensity separately from adherence. |
| A private dermato-gynecology practice may have an atypical referral population | Patients may have more severe, persistent, or complex VLS than patients managed in general pediatrics, dermatology, or public clinics. This can affect both persistence estimates and structural complication rates. | Use multicenter recruitment across primary, secondary, and tertiary settings, including public clinics and community dermatology practices. |
A concise journal-club comment
The long follow-up and incorporation of physical examination are important strengths. However, the main result should be interpreted as an association, not proof that adherence itself causes remission or prevents scarring. The very small fully adherent subgroup, reliance on broad self-reported adherence categories, participant attrition, and likely confounding by baseline disease severity make causal claims premature.
Useful final audience question for the authors
For the VLS study:
Did you compare baseline severity, treatment history, and demographic characteristics between the 68 included participants and the 67 screened patients not included? This would help determine how much selection and attrition may have influenced the reported remission and adherence estimates.
For the LUMBAR study:
Before recommending no imaging for strictly lateral lesions, can the proposed midline criterion be validated prospectively in an unselected cohort of infants with lower-body segmental hemangiomas, with uniform screening for occult LUMBAR anomalies?
Tell me what other studies on similar topic had done different that lacked in these studies insividually
Reading File
Reading File
Searching PubMed
LUMBAR syndrome infantile hemangioma screening
Searching PubMed
pediatric vulvar lichen sclerosus follow-up cohort
Searching the Web
pediatric vulvar lichen sclerosus long-term cohort Boero Kherlopian Fischer adherence
Searching the Web
LUMBAR syndrome segmental infantile hemangioma prospective cohort screening MRI study
Finding Sources
Searching PubMed
"Childhood-onset vulvar lichen sclerosus" puberty
Searching PubMed
"pediatric vulvar lichen sclerosus" "structural abnormalities" corticosteroid
Reading Article
PMID: 20828712
Reading Web Page
https://www.ovid.com/jnls/jped/abstract/00005186-201011000-0…
Reading Article
PMID: 20199450
Reading Article
PMID: 34738263
Reading Article
PMID: 12380449
Reading Article
PMID: 33989804
Reading Article
PMID: 25477189
Reading Article
PMID: 34281089
1. Metry et al. on midline segmental infantile hemangioma and LUMBAR syndrome
There are few directly comparable LUMBAR studies, because LUMBAR is rare. The most methodologically useful comparison is the multicenter prospective study by Drolet et al., 2010, which examined lumbosacral hemangiomas as markers of occult spinal anomalies.
| What other studies did differently | What Metry et al. lacked | Why it is relevant |
|---|---|---|
| Prospective recruitment of at-risk infants | Metry et al. retrospectively reviewed photographs of patients already known to have LUMBAR syndrome. | This is a major difference. A prospective approach reduces publication bias and permits identification of the denominator: how many infants with a particular hemangioma phenotype actually have an anomaly. |
| Multicenter design | Metry et al. combined published reports and a small number of unpublished cases, rather than prospectively enrolling consecutive patients across centers. | Drolet et al. recruited through nine Hemangioma Investigator Group sites, which improves representativeness and reduces dependence on memorable or severe published cases. |
| Objective, uniform imaging outcome | In Metry et al., the key outcome was the morphology and midline relationship of hemangiomas in established LUMBAR cases. It did not systematically image a cohort of lateral and midline lesions to compare anomaly rates. | Drolet et al. performed MRI to identify spinal anomalies. This allowed calculation of a positive predictive value and evaluation of ultrasound performance. Their study found spinal anomalies in 21 of 41 imaged participants, with a positive predictive value of 51.2% for spinal dysraphism. |
| Comparison of clinical risk modifiers | Metry et al. did not formally analyze the independent predictive contribution of lesion size, ulceration, other cutaneous markers, or exact anatomic subsite. | Drolet et al. examined ulceration and additional cutaneous anomalies as risk factors. A LUMBAR study could similarly test whether midline involvement adds predictive value beyond those features. |
| Estimation with confidence intervals | Metry et al. reported that no strictly lateral lesion occurred among their 91 photo-confirmed LUMBAR cases, but did not provide a confidence interval for the residual risk. | A “zero-event” observation does not prove zero risk. A 95% upper confidence limit would show how much uncertainty remains. Using the rule of three, 0 events in 91 cases still permits an approximate upper 95% bound of about 3.3% within a comparable population. |
Important nuance: Drolet et al. studied occult spinal anomalies, not the whole LUMBAR phenotype. Therefore, it cannot directly validate Metry et al.’s conclusion about all LUMBAR-associated anomalies. But it demonstrates the stronger study design needed: prospective recruitment, standardized imaging, and a defined denominator. Drolet et al. prospective study used MRI-based ascertainment and a multicenter prospective cohort.
Journal-club phrasing
Earlier prospective multicenter work evaluated infants with a defined hemangioma phenotype using uniform MRI and therefore could estimate predictive value. In contrast, this study starts with known LUMBAR cases and works backward from photographs. It identifies a potentially useful clinical pattern, but it cannot yet validate midline involvement as a stand-alone screening or rule-out test.
How a better LUMBAR study would look
- Recruit consecutive infants with lumbosacral, sacrococcygeal, pelvic, perineal, gluteal, or lower-limb segmental IH.
- Include midline, paramedian, and strictly lateral lesions.
- Use pre-specified, standardized photographs and blinded morphology classification.
- Perform a uniform evaluation for spinal, renal, pelvic, urogenital, anorectal, bony, and arterial anomalies.
- Report sensitivity, specificity, positive and negative predictive values, likelihood ratios, and confidence intervals for midline involvement.
- Follow children with initially negative screening to identify delayed presentations.
A current prospective screening study is attempting to define a lesion-size threshold for LUMBAR-associated anomalies, which further illustrates that optimal screening criteria remain unsettled. Ongoing screening study
2. Savage et al. on long-term outcomes of prepubertal vulvar lichen sclerosus
This study has stronger long-term follow-up than many earlier papers, but several related studies did certain things more clearly or more rigorously.
| Comparable study | What it did differently | What Savage et al. lacked or could have done better |
|---|---|---|
| Smith and Fischer, 2009: prospective case series | Followed a defined cohort prospectively into adolescence/young adulthood and documented ongoing active disease, need for maintenance therapy, and architectural change. | Savage et al. was mixed retrospective-prospective. Prospective follow-up from diagnosis provides more reliable timing of relapse, treatment exposure, symptoms, and disease activity, with less recall bias. |
| Powell and Wojnarowska, 2002: post-pubertal course study | Specifically documented both symptoms and physical examination findings after puberty. It showed that symptoms can improve while objective signs persist. | Savage et al. partly used questionnaire-defined disease activity and not all patients had uniform clinical review or comparable photographs. This can misclassify “remission,” particularly when symptoms are absent but scarring or subtle active signs remain. |
| Casey et al., 2015: treatment cohort of 72 children | Used a more clearly defined treatment exposure: daily clobetasol propionate 0.05% for 3 months, with follow-up of 4 to 8 years, and compared it with a retrospective cohort treated with lower-potency corticosteroids. | Savage et al. described a practice protocol but did not fully standardize or quantify actual treatment exposure over 19 years. It is difficult to separate the effect of adherence from steroid potency, cumulative dose, maintenance regimen, treatment changes, or baseline severity. |
| Kherlopian and Fischer, 2022: adherence and structural outcome cohort | Included 118 cases, compared 84 “compliant” with 34 “partially compliant” patients, and reported risk ratios with 95% confidence intervals for structural abnormalities. | Savage et al. had only 13 fully adherent patients. Its remission estimate of 92.3% is therefore based on 12 of 13 patients and is statistically imprecise. It would have been stronger to report risk ratios or adjusted estimates with confidence intervals, rather than emphasizing percentages alone. |
| Kammire et al., 2021: disease-course survey | Studied treatment type and duration, current symptoms, current maintenance use, and duration of remission. | Savage et al. had the advantage of dermatologist assessments, but could have more explicitly analysed time to remission, duration of treatment-free remission, and time to relapse using survival analysis rather than classifying the most recent status only. |
| Systematic reviews of pediatric VLS | Emphasize heterogeneity in definitions of remission, follow-up periods, treatment regimens, and outcome measures across pediatric VLS research. | Savage et al. could have used a pre-specified sustained-remission definition, for example: no symptoms, no clinician-detected active signs, and no treatment for a defined minimum duration. Their current definition may capture a temporary inactive phase rather than durable remission. |
The 2009 prospective series found that 9 of 12 patients with active disease at puberty continued to need maintenance therapy after menarche, and it documented architectural change. Smith and Fischer therefore supports the authors’ point that VLS often persists, while using a more clearly prospective follow-up strategy.
The 2002 study similarly used clinical examination after puberty and found physical signs in 16 of 21 patients despite symptomatic improvement in many. Powell and Wojnarowska highlights why symptoms alone are insufficient for defining remission.
For treatment effect, Casey et al. used a more explicit corticosteroid regimen and longer planned treatment follow-up. That makes its treatment-outcome comparison easier to interpret than a retrospective classification of adherence to variable real-world regimens.
For adherence and structural outcomes, Kherlopian and Fischer had a larger cohort and quantified associations: partially compliant patients had higher risks of structural abnormalities. However, it too was retrospective and used nonstandardized adherence categories, so neither paper proves that adherence itself causes the better outcomes.
Journal-club phrasing
Several prior VLS studies either followed patients prospectively, used a more standardized corticosteroid regimen, documented objective examination findings after puberty, or reported effect estimates with confidence intervals. Savage et al. adds long follow-up and quality-of-life assessment, but its causal inference is weakened by retrospective adherence assessment, a very small fully adherent subgroup, treatment heterogeneity, and nonuniform outcome assessment.
One balanced comment to make in discussion
Compared with prior literature, this study contributes longer follow-up plus quality-of-life and examination data. However, the studies that are strongest for treatment inference either prospectively followed patients or defined treatment exposure more consistently. Therefore, the finding should be stated as: adherence was associated with remission and fewer structural changes, rather than adherence was proven to cause those outcomes.
Make a paragraph out of it- Comparison of clinical risk modifiers Metry et al. did not formally analyze the independent predictive contribution of lesion size, ulceration, other cutaneous markers, or exact anatomic subsite. Drolet et al. examined ulceration and additional cutaneous anomalies as risk factors. A LUMBAR study could similarly test whether midline involvement adds predictive value beyond those features.
Unlike Drolet et al., Metry et al. did not formally assess whether other clinical factors, such as lesion size, ulceration, associated cutaneous markers, or the precise lumbosacral/pelvic subsite, independently influence the risk of LUMBAR-associated anomalies. Drolet et al. showed that ulceration and additional cutaneous anomalies were associated with a higher likelihood of spinal abnormalities. Therefore, a stronger LUMBAR study should evaluate midline involvement alongside these features in a multivariable analysis to determine whether midline involvement provides independent predictive value or simply occurs with other high-risk lesion characteristics.
Kherlopian and Fischer, 2022: adherence and structural outcome cohort Included 118 cases, compared 84 “compliant” with 34 “partially compliant” patients, and reported risk ratios with 95% confidence intervals for structural abnormalities. Savage et al. had only 13 fully adherent patients. Its remission estimate of 92.3% is therefore based on 12 of 13 patients and is statistically imprecise. It would have been stronger to report risk ratios or adjusted estimates with confidence intervals, rather than emphasizing percentages alone. What was the kherlopion study about?
Reading File
Reading Article
PMID: 34738263
Searching the Web
Kherlopian Fischer 2022 pediatric vulvar lichen sclerosus compliance structural abnormalities study
Searching PubMed
pediatric vulvar lichen sclerosus adherence
Kherlopian and Fischer’s 2022 study was a retrospective cohort study asking whether regular use of topical corticosteroids reduces permanent vulvar scarring and anatomical changes in girls with vulvar lichen sclerosus.
They reviewed 118 pediatric cases diagnosed between 2004 and 2021. Patients were categorized as compliant with prescribed topical corticosteroid therapy (84 girls) or partially compliant (34 girls). Their main outcomes were permanent structural abnormalities, particularly clitoral phimosis and small, fused, or resorbed labia minora, including abnormalities that persisted beyond menarche.
They found that partially compliant patients had substantially greater risk of complications than compliant patients:
- Risk of any vulvar structural abnormality was 5.76 times higher (95% CI 2.96-11.3).
- Risk of clitoral phimosis was 21.2 times higher (95% CI 5.23-85.9).
- Among girls with prepubertal-onset disease, the risk of a structural abnormality persisting after menarche was 3.54 times higher (95% CI 1.75-7.17) in partially compliant patients.
So, its main message was that consistent topical corticosteroid treatment was associated with less permanent scarring and anatomical distortion. It was not primarily a study of remission rate or quality of life, unlike Savage et al.
However, it still had important weaknesses: it was retrospective, follow-up length varied, there was no untreated control group, and adherence categories were not standardized. Thus, it shows a strong association, but cannot definitively prove that adherence alone caused the lower risk of structural changes. The study is Kherlopian and Fischer, Pediatric Dermatology 2022, PMID: 34738263.
This is a shared conversation. Sign in to Orris to start your own chat.