Children’s Hospital of Eastern Ontario, University of Ottawa
Introduction: Large population-based studies report 30-40% of boys with torsion result in orchiectomy. Few studies evaluate rates of testicular atrophy in patients who do not undergo surgical removal. Our goal was to evaluate atrophy rates as well as clinical, radiographic and surgical predictors of testicular atrophy in children post-orchidopexy for testicular torsion.
Methods: Electronic health records with ICD-9 diagnosis of testicular torsion and ICD-10 procedure of urgent detorsion with orchidopexy in pediatric patients at a single pediatric hospital 2010 to 2020 were reviewed. Patients without preoperative ultrasound were excluded. Patient age, pain duration, TWIST score, ultrasound(US) findings, operative testicular colour, and post-operative US were reviewed. Predictors of testicular atrophy (> 50%) were assessed with Fisher’s exact test, Mann-Whitney U and Student’s t-test, Chi-square. Multivariate logistic regression (backward stepwise method) with assessment goodness-of-fit of the model by Hosmer-Lemeshow test and Nagelkerke R2.
Results: 179/189 patients underwent orchiopexy for testicular torsion with post-operative follow-up were included. Overall, 22 (12.3%) patients had testicular atrophy and 157 (87.7%) had viable testes. Patients with testicular atrophy were younger than those with viable testes (10.6 vs. 14 years, p<0.001). Median hours of symptoms were significantly longer in the atrophy group (24 vs. 4.8 hrs p< 0.001). TWIST score was greater in the atrophy group (5 vs. 4, p=0.046). Use of ultrasound, time from triage to surgery was not different in both groups. Significant US predictors of testicular atrophy included presence of scrotal wall edema (OR 2.11, p=0.026) and heterogeneous testicular parenchyma (OR 2.85, p<0.001). Surgical predictors of atrophy including previously validated colour wheel was reported in 165 cases (Chi Square p<0.001). All cases with post-detorsion black or burgundy colour underwent orchiectomy and pathologic confirmation of non-viability and were not included in study denominator. Testicular atrophy was seen in 2/3 purple (66.7%), 1/9 blue (11.2%), 13/55 mottled pink (23.2%) and 3/96 (3.1%) of pink. The multivariate logistic regression model was statistically significant, Chi-square = 46.562, p<0.001. Hosmer-Lemeshow test p-value for goodness-of-fit was 0.884. The model explained 70.5% (Nagelkerke R2) of variance in atrophic testis and correctly classified 94.7% of cases. Heterogeneous testicular parenchyma on pre-operative ultrasound was 95.34 times more likely to exhibit atrophic testis on follow up. Each additional hour of symptom duration increased odds of atrophy testis by 1.074.
Conclusions: Preoperative factors including symptom duration and ultrasound heterogeneity were most significant factors to predict testicular atrophy. Non-black/hemorrhagic testicular colour has some viable cases in all subjective surgeon assessment of colours, emphasizing the importance of review of preoperative factors in surgical decision making and preoperative patient counselling.
.jpg)
.jpg)
