BackgroundThe use of generative large language models (LLMs) in healthcare is rapidly increasing, offering easier access to medical information. However, comprehensive data on their multilingual accuracy and the reliability of cited scientific references remain limited. This study aimed to compare the response accuracy and reference quality of three free LLMs (ChatGPT, Google Gemini, DeepSeek) in Turkish and English using common pregnancy-related questions.MethodsIn this comparative observational study, 14 frequently asked pregnancy questions were posed to each LLM in Turkish and English, requesting responses supported by up-to-date scientific web sources. Answers were evaluated blindly by obstetricians and gynaecologists for accuracy. References were independently assessed for reliability, scientific validity, and accessibility. Statistical analyses were performed.ResultsLanguage and model infrastructure significantly influenced performance. Google Gemini and DeepSeek provided more accurate responses in English than in Turkish (p

Background/Objectives: This study aimed to evaluate the association between the clinical parameters of vaginal infection—specifically the presence, type, number of concurrent symptoms, and recurrence frequency—and cervical cytology findings, including inflammation severity, Candida, bacterial vaginosis, cellular abnormalities, and Human Papillomavirus (HPV) positivity. Methods: This prospective, cross-sectional, observational study included 458 women attending our gynecology outpatient clinic for Pap smear screening. Vaginal symptoms were documented through face-to-face interviews using a structured data collection form. Cervical samples were evaluated via liquid-based cytology by a single, experienced cytopathologist, who was blinded to the clinical data; cellular abnormalities, the degree of inflammation and cytomorphological findings indicative of infection were reported. HPV analysis was performed on the 218 women for whom results were available. Chi-square and trend chi-square tests were used in the statistical analysis. Results: No significant association was found between the clinical parameters of vaginal symptoms—specifically presence, concurrency, and recurrence frequency—and cytological abnormalities, HPV positivity and bacterial vaginosis (p > 0.05). In contrast, the prevalence of moderate-to-severe inflammation was significantly higher in women with vaginal discharge and a greater symptom burden (p < 0.05). Pruritus, dysuria, and vaginal burning were significantly associated with Candida positivity (p < 0.05). However, no significant association was found between the severity of cervical inflammation and abnormal cytology or HPV positivity (p > 0.05). In multivariable logistic regression analyses, neither symptom burden nor cervical inflammation severity was independently associated with abnormal cytology or HPV positivity. Conclusions: Our findings indicate that vaginal symptoms and the severity of cervical inflammation may not serve as definitive or independent discriminatory markers for cellular abnormalities or HPV positivity in this context. Nevertheless, specific symptom patterns may assist clinicians in evaluating localized infectious processes. Consequently, while standard cytological and molecular protocols remain essential for oncogenic screening, evaluating the overall symptom burden provides clinicians with a valuable framework for identifying benign dysbiotic and inflammatory processes. These findings remained consistent after adjustment for major clinical confounders.

Tubo-ovarian abscess (TOA), a serious complication of pelvic inflammatory disease, poses a serious risk to sexually active women, characterized by high rates of morbidity and mortality. Despite antibiotic therapy being the primary treatment for non-ruptured, stable TOA cases, failure rates of medical management remain as high as 30%. This study investigates the potential of systemic inflammatory markers, particularly the systemic immune-inflammation index (SII), in predicting the necessity for surgical intervention following the failure of medical treatment in TOA patients. A retrospective analysis was conducted on 101 patients diagnosed with TOA at the Bağcılar Training and Research Hospital’s Department of Obstetrics and Gynecology between January 2019 and June 2023. Data collected included demographic details, obstetric history, chronic conditions, clinical history (including intrauterine interventions), and preoperative admission laboratory parameters (CBC, CRP, Neutrophil-to-Lymphocyte Ratio (NLR), and Systemic Immune-Inflammation Index (SII)). Patients were categorized based on treatment outcomes: successful medical treatment versus surgical intervention. A multivariable logistic regression analysis was conducted to assess the predictive value of these markers and to identify independent predictors of medical treatment failure. The surgical group had significantly higher SII, NLR, and CRP levels compared to the medically managed group. An SII cut-off value of 1750 was a significant predictor of medical treatment failure, with an area under the curve (AUC) of 0.749 (95% CI: 0.649–0.850), yielding a sensitivity of 73.2% and specificity of 76.7%. Similarly, NLR demonstrated predictive value with a cut-off value of 4.3, achieving an AUC of 0.694 (95% CI: 0.590–0.798), with 80.5% sensitivity and 58.3% specificity. In the multivariable analysis, SII (OR: 1.002), NLR (OR: 0.706), length of hospital stay (OR: 1.181), and history of curettage (OR: 7.018) were identified as independent predictors of medical treatment failure. Although the mean abscess diameter was larger in the surgical group (64.0 mm vs 57.5 mm), this difference did not reach statistical significance (p=0.079). Systemic inflammatory markers –particularly, the Systemic Immune-Inflammation Index (SII)– appear to be robust, cost-effective, and readily available tools for predicting medical treatment failure in patients with TOA. Predictive performance is enhanced when SII is interpreted in conjunction with NLR and relevant clinical history, especially prior curettage. Integrating these independent predictors into clinical decision-making may facilitate earlier identification of high-risk patients and expedite escalation to surgical intervention, thereby minimizing delays and improving patient outcomes.

Dietary habits are increasingly recognized as an important modifiable risk factor in the management and prevention of polyendocrine metabolic ovarian syndrome (PMOS). However, evidence regarding the relationship between the Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) diet and PMOS is limited. The aim of this study is to investigate the relationship between the MIND diet score and PMOS, as well as its various phenotypes. This case-control study was conducted on 367 women (161 cases, 206 controls) who attended a gynecology and obstetrics clinic. PMOS and its phenotypes were determined according to the Rotterdam criteria. Data were collected face-to-face using the Demographic Information Form, anthropometric measurements, and the MIND Diet Adherence Scale. Multivariate logistic regression was used to estimate the relationship between the MIND diet and PMOS and its phenotypes. The significance level was p < 0.05. The median age of the case and control groups was 22.00 years. Phenotype A was the most prevalent in the case group (43.5%). In model 4, which included all confounding factors, it was found that the likelihood of PMOS was significantly lower in the Q4 group who had the highest MIND diet score compared to the Q1 group (AOR:0.166; 95% CI: 0.073-0.378). In terms of phenotypes, the likelihood of having phenotypes A, C, and D was significantly lower in the Q4 group (p < 0.05); however, no statistically significant relationship was found for phenotype B (p > 0.05). This case-control study found that a higher MIND diet score was associated with a lower likelihood of PMOS and of phenotypes A, C, and D.

Amaç: Toksoplazma gondii, Rubella, Sitomegalovirüs (CMV) gibi enfeksiyöz ajanlar gebelik döneminde intrauterin enfeksiyonlara yol açarak fetal morbidite, mortalite ve anomalilere sebep olabilmektedir. Bu çalışmada Toksoplazma gondii, Rubella, CMV seroprevalanslarının saptanması amaçlanmıştır. Gereç ve Yöntem: Tanımlayıcı retrospektif tipteki araştırmada bir eğitim ve araştırma hastanesi kadın hastalıkları ve doğum polikliniğine 01/01/2017- 01/07/2021 tarihleri arasında gebelik takibi amacıyla başvurarak Toksoplazma gondii, Rubella ve CMV taraması yapılmış olan gebelerin antikor düzeyleri hastane kayıtlarından retrospektif olarak elde edilmiş, yıllara ve yaş gruplarına göre seropravalans değerleri hesaplanmıştır. Bulgular: Bu çalışmada Toksoplasma gondii için 4897, Rubella için 4903 ve CMV için 4600 gebenin test sonuçları değerlendirilmiştir. Sırasıyla Toksoplasma gondii IgM, IgG %1.1, %24.5, Rubella IgM, IgG pozitifliği %0,4, %92.3, CMV IgM, IgG pozitifliği %1.9, %98.6’dır. Toksoplazma IgM(-) saptananların oranı 25-29 yaş grubundaki gebelerde diğer yaş kategorilerine göre yüksektir. 30 yaş ve üzeri gebelerde 25-29 yaş grubuna göre, 25-29 yaş grubu gebelerde 24 yaş ve altındaki gebelere göre Toksoplazma IgG (-) saptananların oranı düşüktür (p0.05). Sonuç: Türkiye’de yapılan çalışmalara göre Toksoplazma IgM (+), IgG (+), Rubella IgG (+) seroprevalansları düşüktür, CMV IgG seropozitifliği ise Türkiye’de yapılan çalışmalara benzerdir. 25-29 yaş grubundaki gebelerde Toksoplazma IgM (-) oranı diğer yaş gruplarına göre yüksektir. Gebelerin yaşının artmasıyla Toksoplazma IgG (-) oranı azalmaktadır. Sitomegalovirüs (CMV) için taramanın cost effektif olmadığı söylenebilir, Toksoplazma gondii ve Rubella için ise taramaların yapılması önerilir.

Amaç: Antenatal dönemde yenidoğanda ortaya çıkan fiziksel yaralanmalar “doğum travması” olarak belirtilmektedir. Klavikula kırıkları yenidoğanlarda kemiğe bağlı en sık görülen doğum travmasıdır. Çalışmamızda yenidoğan muayenesinde klavikula kırığı saptanan olguların değerlendirilmesi amaçlandı. Gereç ve Yöntemler: Hastanemizde 1 Ocak 2022 ile 1 Ağustos 2024 tarihleri arasında doğan bebeklerin dosyaları retrospektif olarak incelendi. Çalışmamızda belirtilen 2,5 yıllık zaman aralığında hastanemizde doğum sayısının 2756 olduğu (2022:1185,2023:1044,2024:527) tespit edildi. Bu doğumların içerisinde klavikula kırığı tespit edilen 31 yenidoğan hasta çalışmaya dahil edildi. Bulgular: Toplam doğumların %42,6’sı normal spontan vajinal yol ile olurken %57,4’ünün sezaryen yöntemi ile olduğu tespit edildi. Çalışmamızda olguların kız/erkek oranı 13/18 idi. Olguların anne yaşı ortalaması 26,64 ± 4,61 gebelik haftası ortalaması 39,08 ± 1,04 idi. Doğum şekli değerlendirildiğinde normal vajinal doğum ile doğan 29 (%93,5), sezaryen ile doğan 2 (%6,5) olgu saptandı. Çalışmamızda doğum tartısı en az olan olgu 2770 gr iken en fazla doğum tartısına sahip olgu 4640 gr’dı. Olguların doğum ağırlığı ortalaması 3462,25 ± 453,11 gr saptandı. Klavikula kırığı saptanan olgularımızın %93,5’i (n:29) vajinal yol ile dünyaya gelirken, %6,5 ‘inin (n:2) ise sezaryen ile doğduğu saptandı. Sonuç: Çalışmamızda tüm doğumların %1,1’inde klavikula kırığı saptandı. Ayrıntılı yenidoğan muayenesinde klavikula kırığı şüphesi olan hastalarda direk grafi ile tanı konmaktadır. Genellikle konservatif yaklaşım ile sekelsiz iyileşme olmaktadır. Doğum travmasının şiddetine bağlı gelişen sinir hasarı varlığında uzun dönem takip ve komplikasyonlar görülebilmektedir.