Prediction score for recurrent adnexal torsion in women with a previous adnexal torsion

Can Natural Language Processing Improve Adnexal Torsion Predictions?

STUDY OBJECTIVE

To create a decision support tool based on machine learning algorithms and natural language processing (NLP) technology, to augment clinicians' ability to predict cases of suspected adnexal torsion.

DESIGN

Retrospective cohort study SETTING: Gynecology department, university-affiliated teaching medical center, 2014-2022.

PATIENTS

This study assessed risk-factors for adnexal torsion among women managed surgically for suspected adnexal torsion based on clinical and sonographic data.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The dataset included demographic, clinical, sonographic, and surgical information obtained from electronic medical records. NLP was used to extract insights from unstructured free text and unlock them for automated reasoning. The machine learning model was a CatBoost classifier that utilizes gradient boosting on decision trees. The study cohort included 433 women who met inclusion criteria and underwent laparoscopy. Among them, 320 (74%) had adnexal torsion diagnosed during laparoscopy, and 113 (26%) did not. The model developed improved prediction of adnexal torsion to 84%, with a recall of 95%. The model ranked several parameters as important for prediction. Age, difference in size between ovaries, and the size of each ovary were the most significant. The precision for the "no torsion" class was 77%, with a recall of 45%.

CONCLUSIONS

Using machine learning algorithms and NLP technology as a decision-support tool for the diagnosis of adnexal torsion is feasible. It improved true prediction of adnexal torsion to 84% and decreased cases of unnecessary laparoscopy.

Diagnostic accuracy of ultrasound signs for detecting adnexal torsion: systematic review and meta-analysis

OBJECTIVE

To evaluate the diagnostic accuracy of different ultrasound signs for diagnosing adnexal torsion, using surgery as the reference standard.

METHODS

This was a systematic review and meta-analysis of studies published between January 1990 and November 2021 evaluating ovarian edema, adnexal mass, ovarian Doppler flow findings, the whirlpool sign and pelvic fluid as ultrasound signs (index tests) for detecting adnexal torsion, using surgical findings as the reference standard. The search for studies was performed in PubMed/MEDLINE, CINAHL, Scopus, The Cochrane Library, ClinicalTrials.gov and Web of Science databases. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool was used to evaluate the quality of the studies. Pooled sensitivity, specificity, and positive and negative likelihood ratios were calculated separately, and the post-test probability of adnexal torsion following a positive or negative test was also determined.

RESULTS

The search identified 1267 citations after excluding duplicates. Eighteen studies were ultimately included in the qualitative and quantitative syntheses. Eight studies (809 patients) analyzed the presence of ovarian edema, eight studies (1044 patients) analyzed the presence of an adnexal mass, 14 studies (1742 patients) analyzed ovarian Doppler flow, six studies (545 patients) analyzed the whirlpool sign and seven studies (981 patients) analyzed the presence of pelvic fluid as ultrasound signs of adnexal torsion. Overall, the quality of most studies was considered to be moderate or good. However, there was a high risk of bias in the patient-selection and index-text domains (with the exception of the whirlpool sign) in a significant proportion of studies. Pooled sensitivity, specificity, and positive and negative likelihood ratios of each ultrasound sign were 58%, 86%, 4.0 and 0.49 for ovarian edema, 69%, 46%, 1.3 and 0.67 for adnexal mass, 65%, 91%, 7.6 and 0.38 for the whirlpool sign, 53%, 95%, 11.0 and 0.49 for ovarian Doppler findings and 55%, 69%, 1.7 and 0.66 for pelvic fluid. Heterogeneity was high for all analyses.

CONCLUSIONS

The presence of an adnexal mass or pelvic fluid have poor diagnostic accuracy as ultrasound signs of adnexal torsion, while the presence of ovarian edema, the whirlpool sign and decreased or absent ovarian Doppler flow have good specificity but moderate sensitivity for detecting adnexal torsion. © 2022 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Rethinking ovary preservation by adnexal torsion reversal in adolescents: a case of delayed diagnosis

Background

This article discusses the management of an adolescent woman with a delayed diagnosis of adnexal torsion (AT) whose ovaries were successfully preserved.

Case presentation

The patient was a 14-year-old female teen admitted with the chief complaint of lower abdominal pain for 3 days and worsening pain for 2 days. Magnetic resonance imaging suggested a high possibility of torsion in the anterosuperior uterine mass and was accompanied by severe ovarian edema, bleeding, and enlargement. Intraoperatively, the left fallopian tube was characterized by thickening and torsion and appeared blackish purple. The left fallopian tube paraovarian cyst was about 20 cm in size, and the left adnexa was twisted 1080° along the left infundibulopelvic ligament (suspensory ligament of the left ovary). The left ovary appeared blackish purple, with an enlarged diameter of about 10 cm. At the request and with the informed consent of the patient’s parents, we preserved the left ovary and removed the left fallopian tube. The results of the endocrine, ultrasound, and tumor marker tests were normal 1 month after surgery. Follicles and blood flow signals seen in ultrasound examinations indirectly proved the successful preservation of the left ovary in the follow-up.

Conclusions

Our attempt to preserve the ovaries in an adolescent with a delayed diagnosis of AT was successful.

Adnexal torsion: a review of diagnosis and management strategies

PURPOSE OF REVIEW

Adnexal torsion remains a diagnostic challenge heavily reliant on high clinical suspicion, with patient presentation and imaging used as adjuncts to aid in its diagnosis. This review summarizes diagnostic and management techniques of adnexal torsion to assist providers when encountering this surgical emergency.

RECENT FINDINGS

Common findings of adnexal torsion include abdominal pain, nausea, vomiting, and adnexal mass or ovarian enlargement. An elevated neutrophil to lymphocyte ratio may be useful for diagnosis. A 'whirlpool' sign, 'follicular ring' sign, enlarged/edematous ovary, and absent Doppler flow to the ovary are highly suggestive of adnexal torsion. Intraoperative visual diagnosis of ovarian death is highly inaccurate, with only 18-20% of ovaries necrotic on pathological examination. Necrotic appearing ovaries have been shown to have follicular activity on ultrasound one year postoperatively.

SUMMARY

Pelvic ultrasound remains the first-line imaging modality. In patients of reproductive age, we recommend performing detorsion with ovarian conservation, even in cases where the tissue appears necrotic, given poor intraoperative diagnostic rates of tissue death. Retention of ovarian function is also reliant on a timely diagnosis and intervention. We emphasize that the risk of ovarian damage/loss outweigh the risk of a diagnostic laparoscopy in patients of reproductive age.