Daily Endocrinology Research Analysis
Analyzed 57 papers and selected 3 impactful papers.
Summary
A pragmatic RCT showed that offering an inpatient oral glucose tolerance test before hospital discharge after gestational diabetes more than tripled postpartum testing completion and increased patient satisfaction. A mechanistic epigenetics study uncovered how a maternal GNAS exon H splice-site variant disrupts imprinting to cause pseudohypoparathyroidism type 1B. A large South African cohort detailed that most gestational diabetes was diagnosed early and that women with HIV had lower odds of GDM, with distinct insulin sensitivity and beta-cell patterns.
Research Themes
- Implementation strategies to improve postpartum diabetes screening
- Epigenetic imprinting mechanisms in endocrine disorders
- Gestational diabetes pathophysiology across early vs late pregnancy and HIV status
Selected Articles
1. Diabetes Testing Immediately Postpartum After Gestational Diabetes Mellitus: A Randomized Controlled Trial.
In this pragmatic RCT of 104 individuals with prior GDM, offering a bedside inpatient 75-g OGTT before discharge raised postpartum testing completion to 92.3% versus 26.9% with standard outpatient testing, and increased patient satisfaction. Dysglycemia (prediabetes/type 2 diabetes) was identified more often in the inpatient-testing arm.
Impact: This trial provides high-quality evidence for an implementation strategy that directly improves postpartum diabetes screening uptake and detection of dysglycemia after GDM.
Clinical Implications: Offer an inpatient OGTT option prior to delivery discharge for individuals with recent GDM to maximize testing completion and earlier diagnosis of prediabetes/type 2 diabetes, aligning with evolving guidelines.
Key Findings
- Postpartum OGTT completion: 92.3% inpatient vs 26.9% outpatient; RR 3.43 (95% CI 2.18–5.40).
- Higher patient satisfaction with inpatient testing (median DTSQ 35 vs 28; P<.001).
- Prediabetes or type 2 diabetes diagnosed in 50.0% inpatient vs 21.4% outpatient (P=.05).
- Postpartum visit attendance exceeded 90% in both arms.
Methodological Strengths
- Pragmatic randomized design with prespecified outcomes and trial registration (NCT05909046).
- High follow-up and postpartum visit completion in both groups, reducing attrition bias.
Limitations
- Nonblinded design and modest sample size may limit precision and generalizability.
- Single-system context and logistics may vary across care settings.
Future Directions: Assess scalability and cost-effectiveness across diverse health systems, and evaluate downstream maternal metabolic outcomes and infant health.
OBJECTIVE: Patient receipt of outpatient postpartum glucose testing after gestational diabetes mellitus (GDM) remains suboptimal. We conducted a randomized controlled trial (RCT) to determine whether the frequency of receiving a postpartum oral glucose tolerance test (OGTT) differed depending on whether the test was planned for the inpatient or outpatient setting. METHODS: DIP (Diabetes Testing Immediately Postpartum) was a pragmatic, nonblinded RCT. Participants were eligible if they received prenatal care and had a GDM diagnosis. After delivery, individuals were randomized in a 1:1 ratio to receive an OGTT either inpatient before delivery discharge (intervention) or outpatient within 12 weeks of birth (standard care). The primary outcome was completion of a fasting 2-hour 75-g OGTT by 12 weeks postpartum or less. Secondary outcomes included participant satisfaction per the adapted DTSQ (Diabetes Treatment Satisfaction Questionnaire) and diagnosis of prediabetes or type 2 diabetes. Based on an estimated baseline testing rate of 52% and a treatment effect of at least 50%, a sample size of 104 participants was required with 80% power and a two-sided α of 0.05. RESULTS: A total of 104 individuals with GDM were enrolled (52 intervention, 52 standard care). Demographics, unmet social needs, and clinical characteristics did not differ according to group assignment. Postpartum visit completion was more than 90% in both groups. The frequency of postpartum diabetes testing was significantly higher in the inpatient group compared with the outpatient group (92.3% vs 26.9%, relative risk 3.43 [95% CI, 2.18-5.40]), as was the median (IQR) score on the DTSQ (35.0 (31.0, 36.0) vs 28.0 (24.0, 32.0), P<.001). Prediabetes or type 2 diabetes was diagnosed among 50.0% in the inpatient group and 21.4% in the outpatient group (P=.05). CONCLUSION: Diabetes testing before delivery discharge for individuals with GDM resulted in more than a threefold higher completion rate of postpartum glucose testing and greater patient satisfaction compared with outpatient testing. This RCT supports recent guidelines that encourage the option of inpatient postpartum diabetes testing before delivery discharge. CLINICAL TRIALS REGISTRATION: ClinicalTrials.gov, NCT05909046.
2. A maternal exon H splice-site variant leading to pseudohypoparathyroidism type 1B with broad methylation defects in GNAS-differentially methylated regions.
Using long-read sequencing and patient-derived iPSCs, the authors show that a maternally inherited exon H splice-site variant disrupts maternal GNAS-H transcription, which precedes and promotes broad methylation defects across GNAS-DMRs characteristic of sporadic PHP1B. This clarifies a causal sequence linking loss of maternal transcription to imprinting failure.
Impact: This work provides mechanistic insight into imprinting establishment in PHP1B and demonstrates the diagnostic utility of long-read sequencing for imprinting disorders.
Clinical Implications: In suspected familial PHP1B with broad GNAS methylation defects, targeted long-read sequencing spanning exon H and adjacent regions should be considered to detect noncoding splice-site variants; mechanistic understanding may refine molecular diagnostics and counseling.
Key Findings
- Long-read sequencing identified a maternally inherited exon H splice-site variant with broad GNAS-DMR methylation defects.
- Patient iPSCs showed loss of GNAS-NESP55 and GNAS-H expression, increased GNAS-AS, and concordant DMR hyper/hypomethylation patterns.
- Demethylation restored biallelic GNAS-NESP55 but not maternal GNAS-H transcription, indicating primary impairment of maternal GNAS-H.
- Amplicon long-read sequencing in 40 sporadic PHP1B cases found variants in this region are infrequent.
Methodological Strengths
- Integration of long-read sequencing with patient-derived iPSC functional assays.
- Use of demethylation perturbation to dissect directionality between transcription and methylation.
Limitations
- Family-based study with limited generalizability; tissue-specific imprinting not fully assessed.
- Causal pathways inferred from iPSC models may not capture all in vivo context.
Future Directions: Expand to multi-family cohorts, assess tissue-specific imprinting in vivo, and develop clinical pipelines incorporating long-read sequencing for suspected imprinting disorders.
BACKGROUND: The GNAS locus produces multiple transcripts, including the maternally derived, GNAS-H, and GNAS-NESP55 and the paternally derived GNAS-XL, GNAS-A/B, and GNAS-AS, all of which are expressed in specific tissues. GNAS-Gsα is biparentally expressed in most tissues and imprinted in several tissues, specifically, the proximal tubules, thyroid, gonads, and pituitary. These imprinted transcripts are regulated by five differentially methylated regions (GNAS-DMRs), and hypomethylation in the GNAS-A/B:TSS-DMR causes resistance to hormones, including parathyroid hormone, leading to pseudohypoparathyroidism type 1B (PHP1B). Sporadic PHP1B shows broad methylation defects in the GNAS-DMRs, whereas most familial PHP1B cases show localized methylation defects at the GNAS-A/B:TSS-DMR and GNAS-AS2:TSS-DMR. We identified an inherited maternal exon H variant that causes a GNAS methylation pattern typically observed in sporadic PHP1B. RESULTS: To determine the underlying genetic cause in this family, we conducted long-read sequencing (LRS). LRS revealed a maternal exon H variant together with methylation defects of CpGs in the GNAS-DMRs. To evaluate the association of the variant with this familial PHP1B, we established patient-derived induced pluripotent stem cells (iPSCs). Reverse-transcription PCR (RT-PCR) and quantitative RT-PCR in patient-derived iPSCs showed no expression of GNAS-NESP55 or GNAS-H and increased expression of GNAS-AS, together with hypermethylation of the GNAS-NESP:TSS-DMR and hypomethylation of the GNAS-AS1:TSS-DMR and GNAS-XL:Ex1-DMR. RNA sequencing revealed no abnormal GNAS transcripts in iPSCs established from patients. To determine whether loss of GNAS-H expression itself causes PHP1B, or whether methylation defects of the GNAS-DMRs, followed by loss of GNAS-H expression, result in PHP1B in this family, we conducted RT-PCR after demethylation treatment in patient-derived iPSCs. We detected biallelic expressed GNAS-NESP55 transcript and only paternally expressed GNAS-H transcript. These findings indicated impaired maternal GNAS-H transcription in the patients, regardless of the methylation levels in the GNAS-DMRs. Furthermore, amplicon LRS spanning the region from exon H to GNAS-AS exon 1 in 40 sporadic PHP1B patients showed that genomic variants in this region are infrequent. CONCLUSION: Loss of maternal GNAS-H transcript precedes the onset of abnormal methylation imprinting at the GNAS-DMRs and contributes to its establishment. We also revealed that LRS is useful for diagnosing and researching imprinting disorders.
3. Gestational diabetes in South African women with HIV and HIV-seronegative women: Results from the ORCHID cohort.
In the ORCHID cohort (n=1,573), 7.9% had GDM, with 65% diagnosed early in pregnancy. Early GDM was associated with the lowest insulin sensitivity and β-cell function indices. Women with HIV had lower adjusted odds of GDM than HIV-seronegative women (aOR 0.57; 95% CI 0.38–0.85).
Impact: This large prospective study clarifies the predominance and pathophysiology of early GDM and quantifies the association between HIV status and GDM risk in an African setting.
Clinical Implications: Consider early pregnancy screening strategies and targeted interventions for impaired insulin sensitivity/β-cell function; HIV status may refine risk stratification but warrants cautious interpretation.
Key Findings
- GDM prevalence was 7.9% overall; 65% of cases were diagnosed early in pregnancy.
- Early GDM showed the lowest Matsuda, Stumvoll, oral disposition index, and glucose sensitivity at enrollment.
- Women with HIV had lower adjusted odds of GDM than HIV-seronegative women (aOR 0.57; 95% CI 0.38–0.85).
Methodological Strengths
- Prospective cohort with standardized 75-g OGTT at two gestational timepoints using WHO criteria.
- Comprehensive assessment of insulin sensitivity and β-cell function using validated indices.
Limitations
- Observational design with potential residual confounding and limited generalizability beyond South Africa.
- Glycemic assessments limited to two timepoints; perinatal outcomes not detailed.
Future Directions: Evaluate longitudinal maternal-fetal outcomes of early vs late GDM and mechanisms underlying differential risk in women with HIV.
PURPOSE: We assessed the prevalence of gestational diabetes (GDM) in South Africa, including early and late GDM, and investigated insulin sensitivity, 1st phase insulin secretion, and beta cell function to understand the underlying pathophysiology of glucose metabolism in early and late pregnancy. We also examined the association between HIV and GDM. METHODS: We enrolled women with HIV (WWH) and HIV-seronegative women at <18 weeks' gestation. A 75g oral glucose tolerance test was administered at enrollment (early) and 32-36 weeks' gestation (late) to diagnose GDM using WHO criteria. Matsuda, Stumvoll, oral disposition index, and glucose sensitivity from Mari models were calculated. Logistic regression models were used to assess the association of HIV with GDM. RESULTS: Among 1573 (n=668 WWH) participants, median age was 28 years, gestation 13 weeks. Overall, 7.9% had GDM (6.7% WWH, 9.1% HIV-seronegative); of these, 65% had early GDM. Women with early GDM had the lowest Matsuda, Stumvoll, DIo, and glucose sensitivity at enrollment compared to those with late GDM and those without GDM. In adjusted analyses, WWH had lower odds of GDM than HIV-seronegative women [adjusted odds ratio: 0.57, 95% Confidence Interval: 0.38-0.85). CONCLUSION: GDM prevalence in South Africa is similar to North America/Europe. Early GDM was diagnosed in a large majority of women who also exhibited features of poorer insulin sensitivity and beta cell function than those diagnosed late in pregnancy or without GDM. WWH had lower GDM risk than HIV-seronegative women. Future studies to understand the implications of early GDM in African populations are warranted.