Twelve-hour fasting compared with expedited oral intake in the initial inpatient management of hyperemesis gravidarum: a randomised trial

Exploring the Limits of Artificial Intelligence for Referencing Scientific Articles

OBJECTIVE

 To evaluate the reliability of three artificial intelligence (AI) chatbots (ChatGPT, Google Bard, and Chatsonic) in generating accurate references from existing obstetric literature.

STUDY DESIGN

 Between mid-March and late April 2023, ChatGPT, Google Bard, and Chatsonic were prompted to provide references for specific obstetrical randomized controlled trials (RCTs) published in 2020. RCTs were considered for inclusion if they were mentioned in a previous article that primarily evaluated RCTs published by the top medical and obstetrics and gynecology journals with the highest impact factors in 2020 as well as RCTs published in a new journal focused on publishing obstetric RCTs. The selection of the three AI models was based on their popularity, performance in natural language processing, and public availability. Data collection involved prompting the AI chatbots to provide references according to a standardized protocol. The primary evaluation metric was the accuracy of each AI model in correctly citing references, including authors, publication title, journal name, and digital object identifier (DOI). Statistical analysis was performed using a permutation test to compare the performance of the AI models.

RESULTS

 Among the 44 RCTs analyzed, Google Bard demonstrated the highest accuracy, correctly citing 13.6% of the requested RCTs, whereas ChatGPT and Chatsonic exhibited lower accuracy rates of 2.4 and 0%, respectively. Google Bard often substantially outperformed Chatsonic and ChatGPT in correctly citing the studied reference components. The majority of references from all AI models studied were noted to provide DOIs for unrelated studies or DOIs that do not exist.

CONCLUSION

 To ensure the reliability of scientific information being disseminated, authors must exercise caution when utilizing AI for scientific writing and literature search. However, despite their limitations, collaborative partnerships between AI systems and researchers have the potential to drive synergistic advancements, leading to improved patient care and outcomes.

KEY POINTS

· AI chatbots often cite scientific articles incorrectly.. · AI chatbots can create false references.. · Responsible AI use in research is vital..

Oral rehydration therapy versus intravenous rehydration therapy in the first 12 h following hospitalization for hyperemesis gravidarum: A randomized controlled trial

OBJECTIVE

To evaluate oral rehydration therapy (ORT) compared with intravenous rehydration therapy (IVT) in the early inpatient management of hyperemesis gravidarum (HG).

METHODS

A total of 124 women hospitalized for HG from February 10, 2021 till January 6, 2023 were randomized to ORT (n = 61) or IVT (n = 63) for an initial 12 h. Inclusion criteria includes women older than 18 years, with a viable intrauterine pregnancy less than 14 weeks at their first hospitalization for HG with ketonuria of at least 2+. Primary outcomes were (1) satisfaction score with allocated intervention, (2) weight change, and (3) ketonuria change at 12 h. Secondary outcomes included vomiting frequency, nausea score, serial vital signs, hematocrit and electrolyte levels at 12 h, deviation from treatment protocol (cross-over therapy), participant recommendation of allocated treatment to a friend, and length of hospital stay.

RESULTS

Primary outcomes of (1) participant satisfaction score (on a 0-10 visual numerical rating scale) was 7 (interquartile range [IQR] 5-8) versus 9 (IQR 8-10), P < 0.001; (2) weight gain was 293 ± 780 g versus 948 ± 758 g, P < 0.001; and (3) ketonuria improvement was 50/61 (82.0%) versus 49/63 (77.8%) (relative risk [RR] 1.05, 95% confidence interval [CI] 0.88-1.26, P = 0.561) for ORT versus IVT, respectively. For secondary outcomes, vomiting frequency was 2.6 ± 2.7 versus 1.1 ± 1.4 episodes (P < 0.001), participant cross-over rate to opposing treatment 20/61 (32.8%) versus 0/63 (0%) (P < 0.001) (in the 12-h study period) and participant recommendation of allocated treatment to a friend rate 24/61 (39.3%) versus 61/63 (96.8%) (RR 0.41, 95% CI 0.30-0.56, P < 0.001) for ORT versus IVT, respectively. By hospital discharge, 31/61 (50.8%) of women allocated to ORT had required IVT. Other secondary outcomes of serial assessments of nausea score and vital signs, hematocrit and electrolyte levels, and length of hospital stay were not different.

CONCLUSIONS

ORT was inferior to IVT in two primary outcomes and three secondary outcomes. Cross-over rate to intravenous therapy from oral therapy was 50.8% by hospital discharge. Intravenous rehydration therapy should remain as first-line rehydration therapy in the early inpatient treatment of HG.

CLINICAL TRIAL REGISTRATION

The present study was registered in ISRCTN registry on December 6, 2020 with trial identification number: ISRCTN 40152556 (https://doi.org/10.1186/ISRCTN40152556). The first participant was recruited on February 10, 2021.

Treatments for hyperemesis gravidarum: A systematic review

INTRODUCTION

Hyperemesis gravidarum affects 0.3%-3% of pregnant women each year and is the leading cause of hospitalization in early pregnancy. Previous systematic reviews of available treatments have found a lack of consistent evidence, and few studies of high quality. Since 2016, no systematic review has been conducted and an up-to date review is requested. In a recent James Lind Alliance collaboration, it was clear that research on effective treatments is a high priority for both patients and clinicians.

MATERIAL AND METHODS

Searches without time limits were performed in the AMED, CINAHL, Cochrane Library, EMBASE, Medline, PsycINFO, and Scopus databases until June 26, 2023. Studies published before October 1, 2014 were identified from the review by O'Donnell et al., 2016. Selection criteria were randomized clinical trials and non-randomized studies of interventions comparing treatment of hyperemesis gravidarum with another treatment or placebo. Outcome variables included were: degree of nausea; vomiting; inability to tolerate oral fluids or food; hospital treatment; health-related quality of life, small-for-gestational-age infant; and preterm birth. Abstracts and full texts were screened, and risk of bias of the studies was assessed independently by two authors. Synthesis without meta-analysis was performed, and certainty of evidence was assessed using the GRADE approach. PROSPERO (CRD42022303150).

RESULTS

Twenty treatments were included in 25 studies with low or moderate risk of bias. The certainty of evidence was very low for all treatments except for acupressure in addition to standard care, which showed a possible moderate decrease in nausea and vomiting, with low certainty of evidence.

CONCLUSIONS

Several scientific knowledge gaps were identified. Studies on treatments for hyperemesis gravidarum are few, and the certainty of evidence for different treatments is either low or very low. To establish more robust evidence, it is essential to use validated scoring systems, the recently established diagnostic criteria, clear descriptions and measurements of core outcomes and to perform larger studies.

Watermelon and dietary advice compared to dietary advice alone following hospitalization for hyperemesis gravidarum: a randomized controlled trial

BACKGROUND

Hyperemesis gravidarum (HG) affects about 2% of pregnancies and is at the severe end of the spectrum of nausea and vomiting of pregnancy. HG causes severe maternal distress and results in adverse pregnancy outcomes long after the condition may have dissipated. Although dietary advice is a common tool in management, trial evidence to base the advice on is lacking.

METHODS

A randomized trial was conducted in a university hospital from May 2019 to December 2020. 128 women at their discharge following hospitalization for HG were randomized: 64 to watermelon and 64 to control arm. Women were randomized to consume watermelon and to heed the advice leaflet or to heed the dietary advice leaflet alone. A personal weighing scale and a weighing protocol were provided to all participants to take home. Primary outcomes were bodyweight change at the end of week 1 and week 2 compared to hospital discharge.

RESULTS

Weight change (kg) at end of week 1, median[interquartile range] -0.05[-0.775 to + 0.50] vs. -0.5[-1.4 to + 0.1] P = 0.014 and to the end of week 2, + 0.25[-0.65 to + 0.975] vs. -0.5[-1.3 to + 0.2] P = 0.001 for watermelon and control arms respectively. After two weeks, HG symptoms assessed by PUQE-24 (Pregnancy-Unique Quantification of Emesis and Nausea over 24 h), appetite assessed by SNAQ (Simplified Nutritional Appetite Questionnaire), wellbeing and satisfaction with allocated intervention NRS (0-10 numerical rating scale) scores, and recommendation of allocated intervention to a friend rate were all significantly better in the watermelon arm. However, rehospitalization for HG and antiemetic usage were not significantly different.

CONCLUSION

Adding watermelon to the diet after hospital discharge for HG improves bodyweight, HG symptoms, appetite, wellbeing and satisfaction.

TRIAL REGISTRATION

This study was registered with the center's Medical Ethics Committee (on 21/05/2019; reference number 2019327-7262) and the ISRCTN on 24/05/2019 with trial identification number: ISRCTN96125404 . First participant was recruited on 31/05/ 2019.

Systematic evidence map of evidence addressing the top 10 priority research questions for hyperemesis gravidarum

OBJECTIVE

Knowledge gaps regarding hyperemesis gravidarum (HG) are substantial. We aimed to systematically identify and map recent evidence addressing the top 10 priority questions for HG, as published in 2021 in a James Lind Alliance Priority Setting Partnership.

DESIGN

Systematic evidence map.

METHODS

We searched MEDLINE and EMBASE on 12 January 2021 and CINAHL on 22 February 2021 with search terms hyperemesis gravidarum, pernicious vomiting in pregnancy and their synonyms. Results were limited to 2009 onwards. Two reviewers independently screened titles and abstracts to assess whether the studies addressed a top 10 priority questions for HG. Differences were discussed until consensus was reached. Publications were allocated to one or more top 10 research questions. Study design was noted, as was patient or public involvement. Two reviewers extracted data synchronously and both cross-checked 10%. Extracted data were imported into EPPI-Reviewer software to create an evidence map.

OUTCOME MEASURES

The number and design of studies in the search yield, displayed per the published 10 priority questions.

RESULTS

Searches returned 4338 results for screening; 406 publications were included in the evidence map. 136 publications addressed multiple questions. Numerous studies address the immediate and long-term outcomes or possible markers for HG (question 8 and 9, respectively 164 and 82 studies). Very few studies seek a possible cure for HG (question 1, 8 studies), preventative treatment (question 4, 2 studies) or how to achieve nutritional requirements of pregnancy (question 10, 17 studies). Case reports/series were most numerous with 125 (30.7%) included. Few qualitative studies (9, 2.2%) were identified. 25 (6.1%) systematic reviews addressed eight questions, or aspects of them. 31 (7.6%) studies included patient involvement.

CONCLUSIONS

There are significant gaps and overlap in the current HG literature addressing priority questions. Researchers and funders should direct their efforts at addressing the gaps in the top 10 questions.

Number and quality of randomized controlled trials in obstetrics published in the top general medical and obstetrics and gynecology journals

BACKGROUND

There has been an increasing number of randomized controlled trials published in obstetrics and maternal-fetal medicine to reduce biases of treatment effect and to provide insights on the cause-effect of the relationship between treatment and outcomes.

OBJECTIVE

This study aimed to identify obstetrical randomized controlled trials published in top weekly general medical journals and monthly obstetrics and gynecology journals, to assess their quality in reporting and identify factors associated with publication in different journals.

STUDY DESIGN

The 4 weekly medical journals with the highest 2019 impact factor (New England Journal of Medicine, The Lancet, The Journal of the American Medical Association, and British Medical Journal), the top 4 monthly obstetrics and gynecology journals with obstetrics-related research (American Journal of Obstetrics & Gynecology, Ultrasound in Obstetrics & Gynecology, Obstetrics & Gynecology, and the British Journal of Obstetrics and Gynaecology), and the American Journal of Obstetrics & Gynecology Maternal-Fetal Medicine were searched for obstetrical randomized controlled trials in the years 2018 to 2020. The primary outcome was the number of obstetrical randomized controlled trials published in the obstetrics and gynecology journals vs the weekly medical journals and the percentage of trials published, overall and per journal. The secondary outcomes included the proportion of positive vs negative trials overall and per journal and the assessment of the study characteristics of published trials, including quality assessment criteria.

RESULTS

Of the 4024 original research articles published in the 9 journals during the 3-year study period, 1221 (30.3%) were randomized controlled trials, with 137 (11.2%) randomized controlled trials being in obstetrics (46 in 2018, 47 in 2019, and 44 studies in 2020). Furthermore, 33 (24.1%) were published in weekly medical journals, and 104 (75.9%) were published in obstetrics and gynecology journals. The percentage of obstetrical randomized controlled trials published ranged from 1.5% to 9.6% per journal. Overall, 34.3% of obstetrical trials were statistically significant or "positive" for the primary outcome. Notably, 24.8% of the trials were retrospectively registered after the enrollment of the first study patient. Trials published in the 4 weekly medical journals enrolled significantly more patients (1801 vs 180; P<.001), received more often funding from the federal government (78.8% vs 35.6%; P<.001), and were more likely to be multicenter (90.9% vs 42.3%; P<.001), non-United States based (69.7% vs 49.0%; P=.03), and double blinded (45.5% vs 18.3%; P=.003) than trials published in the obstetrics and gynecology journals. There was no difference in study type (noninferiority vs superiority) and trial quality characteristics, including pretrial registration, ethics approval statement, informed consent statement, and adherence to the Consolidated Standards of Reporting Trials guidelines statement between studies published in weekly medical journals and studies published in obstetrics and gynecology journals.

CONCLUSION

Approximately 45 trials in obstetrics are being published every year in the highest impact journals, with one-fourth being in the weekly medical journals and the remainder in the obstetrics and gynecology journals. Only about a third of published obstetrical trials are positive. Trials published in weekly medical journals are larger, more likely to be funded by the government, multicenter, international, and double blinded. Quality metrics are similar between weekly medical journals and obstetrics and gynecology journals.

Rating of four different foods in women with hyperemesis gravidarum: a randomised controlled trial

OBJECTIVE

To evaluate four foods in women with hyperemesis gravidarum (HG) on their agreeability and tolerability.

DESIGN

Prospective, randomised, within-subject cross-over trial.

SETTING

Single-centre, tertiary, university hospital in Malaysia.

PARTICIPANTS

72 women within 24-hour of first admission for HG who were 18 years or above, with confirmed clinical pregnancy of less than 16 weeks' gestation were recruited and analysed. Women unable to consume food due to extreme symptoms, known taste or swallowing disorder were excluded.

INTERVENTIONS

Each participant chewed and swallowed a small piece of apple, watermelon, cream cracker and white bread in random order and was observed for 10 min after each tasting followed by a 2 min washout for mouth rinsing and data collection.

OUTCOME MEASURES

Primary outcome was food agreeability scored after 10 min using an 11-point 0-10 Visual Numerical Rating Scale (VNRS). Nausea was scored at baseline (prior to tasting) and 2 and 10 min using an 11-point VNRS. Intolerant responses of gagging, heaving and vomiting were recorded.

RESULTS

On agreeability scoring, apple (mean±SD 7.2±2.4) ranked highest followed by watermelon (7.0±2.7) and crackers (6.5±2.6), with white bread ranked lowest (6.0±2.7); Kruskal-Wallis H test, p=0.019. Apple had the lowest mean nausea score and mean rank score, while white bread had the highest at both 2 and 10 min; the Kruskal-Wallis H test showed a significant difference only at 10 min (p=0.019) but not at 2 min (p=0.29) in the ranking analyses. The intolerant (gagged, heaved or vomited) response rates within the 10 min study period were apple 3/72 (4%), watermelon 7/72 (10%), crackers 8/72 (11%) and white bread 12/72 (17%): χ2 test for trend p=0.02.

CONCLUSION

Sweet apple had the highest agreeability score, the lowest nausea severity and intolerance-emesis response rate when tasted by women with HG. White bread consistently performed worst.