Letters generated by ChatGPT: Author who?

AIM: ChatGPT's role in medical writing is a topic of discussion. I experimented whether ChatGPT almost automatically generates Correspondence or Letter addressed to a "translated" article, and thereby wish to arouse discussion regarding ChatGPT use in medical writing. METHODS: I input an English article of mine into ChatGPT, tasking it with generating an English Disagreement Letter (Letter 1). Next, I tasked ChatGPT with translating the manuscript addressed to from English-French-Spanish-German. Then, I once again tasked ChatGPT with generating an English Disagreement Letter addressed to a German manuscript (triplicate translated manuscript) (Letter 2). RESULTS: Letters 1 and 2 are readable and reasonable, shooting the point that the author (myself) felt as the weakness of the article. Letters addressed to French (single translation) and to Spanish (double translation) and longer Letters (corresponding to Letters 1 and 2) are also readable, and thus stand. CONCLUSIONS: Solely based on this experiment, one may be able to write a letter even without understanding the meaning of the paper being addressed, let alone the language of the paper. Although this humble experiment does not conclude anything, I plea for a comprehensive discussion on the implications of these findings.

A clinic doctor transferring a patient as a coauthor of a case report: A preliminary study.

AIM: Contribution to the authorship, including that for case reports, should be appropriately evaluated. I have noticed a scarcity of case reports with clinic doctors listed as coauthors, prompting this investigation. I sought to offer suggestions on the possible reasons for this trend. METHODS: I checked case reports published in the Journal of Obstetrics and Gynaecology Research, the Journal of Medical Case Reports, and the BMJ Case Reports. I identified case reports listing a clinic doctor as a coauthor. I consulted eight professors at Jichi Medical University to ascertain whether case reports from their departments included clinic doctors as coauthors and, if not, the reasons. RESULTS: Among 65 case reports from Japanese institutes published in the Journal of Obstetrics and Gynaecology Research, only one paper lists a clinic doctor as a coauthor. Of 100 and 50 papers published in the Journal of Medical Case Reports and BMJ Case Reports, respectively, none listed a clinic doctor as a coauthor. Six out of eight professors admitted to never considering the idea of including clinic doctors as coauthors. CONCLUSIONS: The scarcity of case reports with clinic doctors as coauthors extends beyond Japanese obstetrics and gynecology, encompassing various specialties worldwide. Center doctors do not think of the idea that a clinic doctor should be a coauthor. A clinic doctor who transferred the patient should be considered as a candidate coauthor depending on his/her scientific contribution. Such an approach could foster an environment encouraging doctors to contribute to academic writing, regardless of their workplace.

Humans-written versus ChatGPT-generated case reports.

AIM: Artificial intelligence, especially ChatGPT, has been used in various aspects of medicine; however, whether ChatGPT can be used in case report writing is unknown. This study aimed to provoke discussion and provide a platform for it. METHODS: I wrote a theoretical case report where cyst aspiration cured a twisted ovarian cyst (Manuscript 4). I tasked ChatGPT with generating case reports by inputting information at three different levels: (1) key message and case profile, (2) addition of key introduction information (including known facts and problems to be solved), and (3) further addition of main discussion points. These inputs resulted in the creation of Manuscripts 1-3, which were subjected to analysis. Manuscript 3, generated by ChatGPT with the deepest information input, was compared with Manuscript 4, the human-authored counterpart. RESULTS: With the least information, Manuscript 1 can stand on its own, but its content is superficial. The more detailed data input, the more readable and reasonable the manuscripts become. A human-written manuscript involves personal experience and viewpoints other than obstetrics-gynecology. CONCLUSIONS: Better input produced more reasonable and readable case reports. Human-written paper, compared with ChatGPT-generated one, can involve "human touch." Whether such human touch enriches the case report awaits further discussion. Whether ChatGPT can be used in case report writing, and if it can, to what extent, should be worthy of further study. I encourage every doctor to form their own stance towards ChatGPT use in medical writing.

Investigation of optimal weight gain during pregnancy: A retrospective analysis of the Japanese perinatal registry database.

AIM: This study aimed to determine the weight gain during pregnancy that minimizes the predicted probability of various perinatal adverse events according to the pre-pregnancy body mass index (BMI) and make recommendations for optimal weight gain in Japan. METHODS: The Japan Society of Obstetrics and Gynecology perinatal database for 2015-2017 was used. From the 719 723 deliveries included in this database, parturients with underlying diseases or missing data were excluded, and 419 114 deliveries were analyzed. A questionnaire survey was also conducted to weigh each perinatal adverse event. For each of the nine outcomes, a restricted cubic spline model was made to estimate the association between the "expected gestational weight gain at 40 weeks" and the outcome risk. RESULTS: Since the classes of medical facilities were generally the same, weights were assigned according to the mean of the questionnaires rather than by the class of the facility. For each pre-pregnancy BMI, the weight gains during pregnancy that minimized the predicted probability of various adverse perinatal events were 12-15, 10-13, 7-10, and upper limit of 5 kg for the underweight, normal-weight, obese 1, and obese ≥2 groups, respectively. CONCLUSIONS: The weight gain during pregnancy that minimizes the predicted probability of various perinatal adverse events according to the pre-pregnancy BMI was established.

Preserving the human voice: A plea for regulating ChatGPT use in Correspondence and Opinion writing.

Correspondence and Opinion provide a 'human voice,' serving as the captivating elements that grab readers' attention and stimulate contemplation. The extent of ChatGPT's involvement in writing raises controversy. We tested ChatGPT's capability to generate readable Correspondence or Opinion, producing an English Opinion from a Japanese abstract. We also tasked ChatGPT with creating Correspondence and Reply for an English→German translated Opinion. Opinion, Correspondence, and Reply output here were found to be readable and reasonable. While preliminary, these results suggest that ChatGPT can generate such articles, prompting serious concern about AI taking over the 'human voice.' Some regulatory measures may be needed.

Bilateral ovarian vein thrombosis without underlying conditions: A case report.

Ovarian vein thrombosis (OVT) usually occurs on the right side in women with underlying conditions such as pelvic diseases and thrombophilia or during the postpartum period. Here, we present a patient with bilateral OVTs without underlying conditions. A 63-year-old woman without remarkable past or family histories complained of mild lower abdominal pain. Computed tomography revealed bilateral OVTs, with the right-sided thrombus nearly progressing into the inferior vena cava. The presence of symptoms and risk of thrombus extension/pulmonary embolism (PE) led us to administer an oral anticoagulant (rivaroxaban) promptly. Her symptom improved gradually and bilateral OVTs disappeared by 3 months without any evidence of PE. The present case suggests that OVT can occur in a woman without underlying conditions, and the use of an anticoagulant may be a treatment option in such a case.

A modified transabdominal cervicoisthmic cerclage with the monofilament thread: Its efficacy and safety for women with extremely short cervix due to cervical conization.

AIM: The study aimed to examine the usefulness of modified transabdominal cervicoisthmic cerclage (TAC) using monofilament thread for the prevention of preterm delivery in women with an extremely short cervix after deep conization. METHODS: We devised a monofilament thread for picking up the seromuscular layer of the site that is slightly cephalad to the internal ostium to prevent injury of the vessels around the uterine cervix. From 2017 to 2020, we performed this modified operation in eight women (nine pregnancies) at 12-16 weeks of gestation with a history of deep cervical conization. RESULTS: A modified TAC was successfully performed in all patients. There was no measurable bleeding, and all patients were discharged without postoperative complications. Their pregnancy courses after the operation were uneventful. Of nine, one patient had premature uterine contractions and underwent cesarean section at 36 weeks (preterm delivery). In the other eight pregnancies, planned cesarean section was performed after 37 weeks of gestation. The median birth weight of the babies was 2996 g (range 2604-3374 g). All patients were discharged on the sixth postoperative day without complications. CONCLUSION: A modified TAC can be safely performed and may prolong pregnancy without adverse events in patients with an extremely short cervix.

Cerebral Aneurysm Rupture during Pregnancy Resulting in Subdural Hematoma without Subarachnoid Hemorrhage.

Subarachnoid hemorrhage is typically present in cerebral aneurysm rupture, whereas acute subdural hematoma without subarachnoid hemorrhage is rare. We herein report a case of cerebral aneurysm rupture during pregnancy resulting in acute subdural hematoma without subarachnoid hemorrhage. A 37-year-old gravida 4 para 3 pregnant woman was admitted for threatened preterm labor at 294/7 weeks of gestation. At 296/7 weeks of gestation (day -14), she developed mild left eye pain, which disappeared within one day. At 316/7 weeks of gestation (day 0), she developed the sudden onset of severe headache and nausea. A neurological examination revealed no abnormal findings, and analgesics ameliorated her headache. At 321/7 weeks of gestation (day 2), after consultations with neurosurgeons, magnetic resonance imaging showed acute subdural hematoma without subarachnoid hemorrhage. Further examinations revealed a cerebral aneurysm. Emergent clipping surgery was performed with the fetus in utero in consideration of the immaturity of the fetus and stable maternal/fetal general conditions. At 356/7 weeks of gestation (day 28), her headache of unknown cause recurred. Considering the maturity of the fetus, the patient underwent cesarean section with good maternal and neonatal outcomes. The absence of subarachnoid hemorrhage does not eliminate cerebral aneurysm rupture.

Low placental growth factor levels and high soluble endoglin levels at 26-31 weeks of gestation precede light placenta with and without relatively heavy infant, respectively: A retrospective cohort study.

AIM: Our aim was to examine whether serum levels of placental growth factor (PlGF) and soluble endoglin (sEng) at 19-25 and 26-31 weeks of gestation were associated with the occurrence of the 9-block categorization of placenta weight (PW) and fetal/placenta ratio (F/P ratio). METHODS: We performed a retrospective cohort study in 1391 women with singleton pregnancy. Serum levels of PlGF and sEng were measured by enzyme immunosorbent assay. A light placenta was defined as PW ZS < -1.28 SD. Based on the PW (light, normal, and heavy) and F/P ratio (relatively heavy, balanced growth, and relatively small), 9-block categorization were performed. Multivariable logistic regression analyses were performed. RESULTS: Low PlGF at 26-31 weeks was an independent risk factor for the birth of infants belonging to Block A (light placenta and relatively heavy infant), after adjusting for prepregnancy body mass index and serum levels of sEng. High sEng at 26-31 weeks was an independent risk factor for the birth of infants belonging to Block D (light placenta and balanced growth of infant), after adjusting for past history of either preeclampsia or gestational hypertension, high pulsatility index of uterine artery flow velocity waveforms in the second trimester, and serum level of PlGF. CONCLUSIONS: Low PlGF levels at 26-31 weeks of gestation may precede a light placenta and relatively heavy infant (Block A), and high sEng levels at 26-31 weeks of gestation may precede a light placenta and balanced growth of infant (Block D).

LncRNA H19-Derived miR-675-5p Accelerates the Invasion of Extravillous Trophoblast Cells by Inhibiting GATA2 and Subsequently Activating Matrix Metalloproteinases.

The invasion of extravillous trophoblast (EVT) cells into the maternal decidua, which plays a crucial role in the establishment of a successful pregnancy, is highly orchestrated by a complex array of regulatory mechanisms. Non-coding RNAs (ncRNAs) that fine-tune gene expression at epigenetic, transcriptional, and post-transcriptional levels are involved in the regulatory mechanisms of EVT cell invasion. However, little is known about the characteristic features of EVT-associated ncRNAs. To elucidate the gene expression profiles of both coding and non-coding transcripts (i.e., mRNAs, long non-coding RNAs (lncRNAs), and microRNAs (miRNAs)) expressed in EVT cells, we performed RNA sequencing analysis of EVT cells isolated from first-trimester placentae. RNA sequencing analysis demonstrated that the lncRNA H19 and its derived miRNA miR-675-5p were enriched in EVT cells. Although miR-675-5p acts as a placental/trophoblast growth suppressor, there is little information on the involvement of miR-675-5p in trophoblast cell invasion. Next, we evaluated a possible role of miR-675-5p in EVT cell invasion using the EVT cell lines HTR-8/SVneo and HChEpC1b; overexpression of miR-675-5p significantly promoted the invasion of both EVT cell lines. The transcription factor gene GATA2 was shown to be a target of miR-675-5p; moreover, small interfering RNA-mediated GATA2 knockdown significantly promoted cell invasion. Furthermore, we identified MMP13 and MMP14 as downstream effectors of miR-675-5p/GATA2-dependent EVT cell invasion. These findings suggest that miR-675-5p-mediated GATA2 inhibition accelerates EVT cell invasion by upregulating matrix metalloproteinases.

Cumulative incidence of Kawasaki disease with cardiac sequelae in Japan.

BACKGROUND: Some patients with Kawasaki disease (KD) develop cardiac sequelae, which increase the risk of subsequent ischemic heart events. Knowing the cumulative incidence (CI) of KD with cardiac sequelae may contribute to developing health policies to prevent subsequent ischemic events in these patients. METHODS: Study participants consisted of 254 984 patients aged 0-9 years with KD who were registered in nationwide surveys in Japan from 1991-2016. We calculated the incidence probabilities by dividing the number of patients with KD aged 0-9 years by the population used in vital statistics of each calendar year. We calculated the cumulative proportion of those not affected by KD, by multiplying each probability in patients aged from age 0-9 years. The CI of KD was obtained by subtracting this value from 1. We also calculated the number of patients in each birth cohort (BC). The same was done to calculate the CI of KD-related cardiac sequelae. RESULTS: The CIBC steadily increased from 0.005067 in males and 0.003668 in females in 1991 to 0.011431 in males and 0.0088253 in females in 2007. The CIBC of KD with cardiac sequelae decreased from 0.000478 in males and 0.000213 in females in 1997 to 0.000339 in males and 0.000169 in females in 2007. CONCLUSION: The increasing CIBC of KD indicates an increased susceptibility to KD in accordance with birth year. The decreasing CIBC of cardiac sequelae suggests the efficacy of KD treatment.

A Nationwide Survey of Obstetric Care Status on Japan's Islands, with Special Reference to Maternal Transport to the Mainland.

Safe obstetric care is a worldwide requirement. In Japan, a reduced number of obstetricians has prompted the centralization of obstetric facilities, and Japan's islands are expected to experience problems in handling deliveries. Although many pregnant women "move" to the mainland at later gestational weeks, "transport" from the island to the mainland may be occasionally needed when disorders manifest before the "move." Other women plan within-island deliveries; however, transport is required when complications arise. Managing delivery- or pregnancy-related problems may differ in transport by the population size of islands. We investigated the following issues in relation to the population size of Japan's islands: 1) How were deliveries handled on islands? 2) How many pregnant women were transported to the mainland? 3) What was the reason for and what affected transport? A total of 142 municipalities were selected to participate in a questionnaire survey, and 108 institutions from 106 municipalities responded. A comparative analysis by island size was performed using 2014 data: small-sized (population < 1,000), mid-sized (1,000 to 5,000), and large-sized (≥ 5,000). The percentage of women transported to the mainland from small-sized islands was significantly higher than that from large-sized islands (6.8 vs. 2.6% of all births in a year, respectively, P < 0.01). Transport was mainly in response to threatened preterm labor. Adverse weather was the most frequent factor affecting transport reliability. Our findings may contribute to a more detailed understanding of the state of obstetric care on Japan's islands.

Usefulness of dynamic computed tomography for diagnosing and evaluating uterine artery pseudoaneurysms in women with late post-partum hemorrhage not complicated by retained products of conception.

AIM: We examined whether critical conditions, which were defined as having hemoglobin (Hb) less than 7.0 g/dL, shock index ≥1.0, or need for transfusion, were associated with the presence of extravasation (EV) on dynamic computed tomography (CT) in women with late post-partum hemorrhage (PPH). METHODS: Forty post-partum women with late PPH without evident retained products of conception performed dynamic CT. Two radiologists retrospectively evaluated dynamic CT, and determined the presence or absence of EV and a sac-like structure within the uterine cavity with enhancement. RESULTS: Ultrasound images were available in 34/40 patients. Color Doppler flow in uterine cavity was evaluated in 33/34 (97%), and all women showed abnormal flow. Of 40 patients, dynamic CT revealed EV in 8 (20%), and a sac-like structure in 30 (75%). Thus, we diagnosed these 38 (95%) as having uterine artery pseudoaneurysm (UAP). Uterine artery embolization was performed in 36/38 diagnosed as having UAP, and in 2/2 patients with an unknown cause of hemorrhage. The incidence rates of critical conditions were significantly increased in PPH women with than without EV on dynamic CT: Hb <7.0 g/dL (62.5 vs 0%, [P < 0.001]), shock index ≥1.0 (50 vs 9.4% [P = 0.020]), and need for transfusion (37.5 vs 0% [P = 0.006]). Abnormal color Doppler flows were observed in all patients with either EV and sac on dynamic CT. CONCLUSION: Dynamic CT was useful for diagnosing UAP, and for evaluating critical conditions, in women with late PPH not complicated by retained products of conception.