Original Report
Diagnostic Performance of a Sonographic Volume and Solid Vascular Tissue Score (VSVTS) for Preoperative Risk Assessment of Pediatric and Adolescent Adnexal Masses

https://doi.org/10.1016/j.jpag.2020.11.017Get rights and content

Abstract

Study Objective

To evaluate the diagnostic performance of a Volume and Solid Vascular Tissue Score (VSVTS) for preoperative risk assessment of pediatric and adolescent adnexal masses.

Design

A retrospective cohort study comprised of all female individuals who presented with an adnexal mass that was managed surgically between April 2011 and March 2016.

Setting

The Hospital for Sick Children (Toronto, Ontario, Canada).

Participants

Female individuals 1–18 years of age who presented to a large tertiary pediatric hospital with an adnexal mass that was managed surgically.

Main Outcome Measures

Main outcome measures included diagnostic performance of the VSVTS for malignancy via sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), negative likelihood ratio (LR−), and receiver operating characteristic area-under-the-curve (AUC) analysis.

Results

A total of 179 masses in 169 subjects were included. The malignancy rate was 10.6%. The AUC for the VSTVS was 0.919. A VSTVS cut-off value of 4 achieved a sensitivity of 79% (95% CI 0.54-0.93), specificity of 88% (95% CI 0.82-0.93), PPV of 0.44 (95% CI 0.33-0.56), NPV of 0.97 (95% CI 0.94-0.99), LR+ of 6.77 (95% CI 4.18-10.97), and LR− of 0.24 (95% CI 0.10-0.57).

Conclusions

A sonographic scoring system based on the volume and presence of solid vascular tissue improves PPV for preoperative risk stratification of adnexal masses in the pediatric and adolescent population compared to existing ultrasound-only approaches. Further prospective research is needed to determine how best to incorporate components of such scoring systems into clinical management algorithms.

Introduction

The incidence of adnexal masses in girls and adolescents is approximately 2.6/100,000 girls per year.1 Up to 10-20% of surgically managed masses are malignant.2 Accurate preoperative risk assessment is critical, as it affects the surgical approach.3 For presumed benign masses that require intervention, a conservative approach with laparoscopic cystectomy is preferred. Malignancies, on the other hand, usually require an open laparotomy with oophorectomy, which has a consequent impact on patient fertility.2,3 Nonetheless, the optimal method for preoperative risk assessment in the pediatric and adolescent population remains elusive.

A previous retrospective study conducted at our institution showed that malignant masses can be preoperatively identified using a threshold of greater than 8 cm in size and presence of “complex” features by ultrasound. These criteria yielded a negative predictive value (NPV) of 100% but a positive predictive value (PPV) of only 37.1%.3 Therefore, relying on such criteria alone would lead to a high proportion of undesirable oophorectomies for benign lesions.

Multiple more-complicated risk assessment algorithms have been studied in attempts to improve the preoperative PPV.2, 3, 4, 5, 6, 7, 8, 9, 10 Most recently, we attempted replication of the Decision Tree System (DTS) proposed by Stankovic et al,4 which in their study demonstrated a preoperative NPV of 99% and PPV of 86%. The results of our replication attempt were less encouraging, as we achieved an NPV of 98% but a PPV of only 30% in our population.4

Concurrently, we hypothesized that a scoring system accounting for the vascularity of adnexal masses as has been introduced in adult algorithms, such as in the IOTA rules and O-RADS,11,12 might improve the PPV. These algorithms incorporate a subjective degree of vascularity in risk assessment, whereas in our experience any solid vascular tissue alone increases malignancy risk. The purpose of this study was therefore to evaluate the diagnostic performance of a preoperative sonographic adnexal mass scoring system based on mass volume and amount of solid vascular tissue in the pediatric and adolescent population.

Section snippets

Patient Population

This is a retrospective study of all female individuals 1-18 years of age presenting with an adnexal mass to The Hospital for Sick Children (Toronto, Ontario, Canada) between April 2011 and March 2016 and was conducted with approval of our institutional Research Ethics Board. Patients who did not receive surgical management or preoperative ultrasound that included spectral Doppler were excluded. The inclusion dates mirror a similar cohort used in previous risk stratification studies from our

Results

In total, 169 patients with 179 masses with surgical management of adnexal masses were identified, with 10.6% of the lesions being malignant. The mean age of the study population was 13.0 ± 3.4 years. 73.7% of the study population was postmenarchal. All pathologic diagnoses are listed in Table 2.

The distribution of pathologic entities per VSVTS is depicted in Table 3. The distribution of pathologic entities per volume score and solid vascular tissue score is available in Supplemental Table 1.

Discussion

In this study, we evaluate the diagnostic performance of a sonographic volume and solid vascular tissue score for preoperative risk stratification of adnexal masses in the pediatric and adolescent population. The major finding of our study is that the addition of a solid vascular tissue score to preoperative assessment shows incremental benefit to PPV and LR + while maintaining reasonable NPV and LR–for prediction of malignancy in pediatric and adolescent adnexal masses.

Multiple

Conclusion

A sonographic scoring system based on the volume and presence of solid vascular tissue improved PPV for preoperative risk stratification of adnexal masses in the pediatric and adolescent population compared to existing ultrasound-only approaches. Further prospective research is needed to determine how best to incorporate components of such scoring systems into clinical management algorithms.

Acknowledgments

The authors would like to acknowledge the support from the Ontasian Imaging Lab (OIL).

Cited by (1)

None of the authors have any conflicts of interest or financial ties to disclose.

The authors report no funding for this work.

View full text