Validation of a clinical risk scoring system, based solely on clinical presentation, for the management of pregnancy of unknown location

Barnhart, Kurt T.; Sammel, Mary D.; Takács, Péter; Chung, Karine; Morse, Christopher B.; O’Brien, Katherine O’Flynn; Allen-Taylor, Lynne; Shaunik, Alka

doi:10.1016/j.fertnstert.2012.09.012

Cited by 9 publications

(8 citation statements)

References 21 publications

(36 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several mathematical models have been developed using clinical presentation 6 or these biochemical markers 7–11 to triage PUL presentations into low or high risk of ectopic pregnancy. These triage protocols, including the M4 9 and M6 10 may be beneficial in guiding management and surveillance.…”

Section: Introductionmentioning

confidence: 99%

How do the M4 and M6 models perform in an Australian pregnancy of unknown location population?

Ooi

Vries

Ludlow

2020

Aust NZ J Obst Gynaeco

View full text Add to dashboard Cite

Background The diagnosis of a pregnancy of unknown location (PUL) is made when there is an elevated serum β human chorionic gonadotropin (βhCG) and no pregnancy on transabdominal and transvaginal ultrasound. Most of these pregnancies end as intra‐uterine pregnancies or unsuccessful pregnancies and can be safely managed expectantly. However, up to 20% of these women will have an ectopic pregnancy. Several mathematical models, including the M4 and M6 protocols, have been developed using biochemical markers to triage PUL presentations. This rationalises numbers of tests and visits made without compromising safety and allowing timely intervention. Aims We aimed to externally validate the M4 and M6 models in an Australian tertiary early pregnancy assessment service (EPAS). Materials and Methods We performed a retrospective single‐centre cohort study across five years. Our study population included all women attending our EPAS with a PUL who had at least two serum βhCG levels and one progesterone level measured. The M4 and M6 models were retrospectively applied. Results Of the 360 women in the study population, there were 26 confirmed ectopic pregnancies (7.2%) and six persisting PULs (2%). The M4 model had a sensitivity and specificity of 72%. The M6P model had a sensitivity of 91% and specificity of 63%. The M6P misclassified two ectopic pregnancies into the low‐risk group, compared with seven in the M4 model. Conclusions The M6P model has the highest sensitivity of the three models and a negative predictive value of 99%. These numbers are comparable to the original United Kingdom population. Further prospective validation is planned.

show abstract

Section: Introductionmentioning

confidence: 99%

How do the M4 and M6 models perform in an Australian pregnancy of unknown location population?

Ooi

Vries

Ludlow

2020

Aust NZ J Obst Gynaeco

View full text Add to dashboard Cite

show abstract

“…The 35 patients included in the study were classified according to Barnhart's “Risk Scoring for Nonviable Pregnancy of Unknown Location” and the risk stratification was performed ( Table 1 ). 10 The 3rd stage shock accompanied by hypotension and tachycardia was defined as hemodynamic instability. 11…”

Section: Methodsmentioning

confidence: 99%

Analysis of ectopic pregnancies admitted to emergency department

Şimşek

2015

Turkish Journal of Emergency Medicine

View full text Add to dashboard Cite

ObjectivesEctopic pregnancy (EP) may cause significant morbidity and mortality. In this study, we aimed to evaluate the demographic characteristics, presence of risk factors and diagnostic parameters of the patient with EP and predicting parameters for ruptured EP.MethodsPatients who presented to emergency department (ED) and diagnosed as EP within one year were included to the study. The demographic characteristics, β-human chorionic gonadotropin (β-HCG) levels, transvaginal ultrasonography (TVUSG) findings, treatment protocols, pathology reports and hemoglobin levels at the time of admission to ED were obtained from patient files and hospital automation system and statistical analysis was performed.ResultsTotal 35 patients were included to the study. The mean age of the patients was 30 ± 5.6 years. Among the patients, 46% had a history of caesarean section (C-section). The complaints of the patients at presentation, their age, gestational week and the β-HCG levels were found to be inefficient in predicting ruptured EPs. TVUSG was found statistically significant in terms of demonstrating ruptures in EP. The ratio of salpingectomies was observed to be higher in the surgical treatment of ruptured EPs.ConclusionsC-section was most frequently seen with EP. There is no absolute diagnostic parameter for predicting ruptured EPs and TVUSG may be a clue for diagnosis. The final diagnosis is made through surgery.

show abstract

“…Attempts in developing models for EP diagnosis were established from a variety of domains. These include the clinical risk factor model, classifying elevated risk group (19) or the risk factor model combined with a single serum hCG (20). While Abbreviations: EP, ectopic pregnancy; PUL, pregnancy of unknown location; ML, machine learning; NNs, neural networks; DT, decision tree; LR, logistic regression; SVM, support vector machines.…”

Section: Introductionmentioning

confidence: 99%

Predictive analytical model for ectopic pregnancy diagnosis: Statistics vs. machine learning

2022

View full text Add to dashboard Cite

ObjectiveEctopic pregnancy (EP) is well known for its critical maternal outcome. Early detection could make the difference between life and death in pregnancy. Our aim was to make a prompt diagnosis before the rupture occur. Thus, the predictive analytical models using both conventional statistics and machine learning (ML) methods were studied.Materials and methodsA retrospective cohort study was conducted on 407 pregnancies with unknown location (PULs): 306 PULs for internal validation and 101 PULs for external validation, randomized with a nested cross-validation technique. Using a set of 22 study features based on clinical factors, serum marker and ultrasound findings from electronic medical records, analyzing with neural networks (NNs), decision tree (DT), support vector machines (SVMs), and a statistical logistic regression (LR). Diagnostic performances were compared with the area under the curve (ROC-AUC), including sensitivity and specificity for decisional use.ResultsComparing model performance (internal validation) to predict EP, LR ranked first, with a mean ROC-AUC ± SD of 0.879 ± 0.010. In testing data (external validation), NNs ranked first, followed closely by LR, SVMs, and DT with average ROC-AUC ± SD of 0.898 ± 0.027, 0.896 ± 0.034, 0.882 ± 0.029, and 0.856 ± 0.033, respectively. For clinical aid, we report sensitivity of mean ± SD in LR: 90.20% ± 3.49%; SVM: 89.79% ± 3.66%; DT: 89.22% ± 4.53%; and NNs: 86.92% ± 3.24%, consecutively. However, specificity ± SD was ranked by NNs, followed by SVMs, LR, and DT, which were 82.02 ± 8.34%, 80.37 ± 5.15%, 79.65% ± 6.01%, and 78.97% ± 4.07%, respectively.ConclusionBoth statistics and the ML model could achieve satisfactory predictions for EP. In model learning, the highest ranked model was LR, showing that EP prediction might possess linear or causal data pattern. However, in new testing data, NNs could overcome statistics. This highlights the potency of ML in solving complicated problems with various patterns, while overcoming generalization error of data.

show abstract

Validation of a clinical risk scoring system, based solely on clinical presentation, for the management of pregnancy of unknown location

Cited by 9 publications

References 21 publications

How do the M4 and M6 models perform in an Australian pregnancy of unknown location population?

How do the M4 and M6 models perform in an Australian pregnancy of unknown location population?

Analysis of ectopic pregnancies admitted to emergency department

Predictive analytical model for ectopic pregnancy diagnosis: Statistics vs. machine learning

Contact Info

Product

Resources

About