Does the source of referral affect outcomes for paediatric testicular torsion?

Abstract

Introduction

This single centre study retrospectively analysed the intraoperative findings relative to source of referral for emergency scrotal explorations performed in a tertiary level paediatric surgery department.

Methods

All patients who underwent emergency scrotal exploration under the care of paediatric surgeons in our unit between April 2008 and April 2016 were identified. Clinical data were obtained from contemporaneous records.

Results

Over the 8-year study period, 662 boys underwent emergency scrotal exploration: 6 (1%) were internal referrals, 294 (44%) attended our emergency department (ED) directly, 271 (41%) were referred from primary care and 91 (14%) were transferred from other hospitals. Excluding procedures in neonates, testicular torsion was present in 100 cases (15%). Testicular detorsion with bilateral 3-point testicular fixation was performed in 66 (66%) and orchidectomy with contralateral fixation in 34 (34%) where the torted testis was non-viable intraoperatively. The orchidectomy rate in the presence of torsion was 23% in ED referrals (12/52), 43% in primary care referrals (12/28) and 50% for transfers (10/20). The difference in rates between ED referrals and patients transferred from other hospitals was significant (p=0.026). There was no significant difference in median age between any of the groups (p=0.10).

Conclusions

Boys undergoing emergency scrotal exploration had a higher orchidectomy rate when transferred from other hospitals to our unit. This difference was statistically significant when compared with boys presenting directly to our ED. This supports advice from The Royal College of Surgeons of England for undertaking paediatric scrotal explorations in the presenting hospital when safe to do so rather than delaying the care of these patients by transferring them to a tertiary paediatric surgical unit.

Sudden onset testicular pain is a common presentation in children and adolescents.¹ Testicular torsion, torsion of the testicular appendage (hydatid of Morgagni) and epididymo-orchitis account for the large majority of cases.^2,3 Other causes include idiopathic scrotal oedema, varicocoele, hydrocoele, trauma, inguinoscrotal hernia, tumour and referred pain (such as from appendicitis).^1,4

It is recognised that there is a 4–8-hour window before significant ischaemic damage occurs in testicular torsion,^5,6 resulting in morphological changes in testicular histopathology and impaired spermatogenesis.^6,7 Delayed exploration leads to increased orchidectomy rates^6,8 and it would therefore follow that patients being transferred from regional hospitals to tertiary paediatric surgical centres are potentially at increased risk of testicular loss.⁹ In the UK, this has resulted in the recommendation that these transfers should occur as an exceptional circumstance and that exploration should be undertaken in the presenting hospital.¹⁰

This single centre study retrospectively analysed the intraoperative findings relative to source of referral for a consecutive series of emergency scrotal explorations performed in a tertiary level paediatric surgery department. The aim was to determine whether boys transferred from local hospitals were at greater risk of testicular loss than those who presented directly to the tertiary unit either via the emergency department (ED) or primary care.

Methods

All patients who underwent emergency scrotal exploration for suspected testicular torsion under the care of paediatric surgery in our unit between April 2008 and April 2016 were identified from the local hospital theatre system. Clinical data were obtained from contemporaneous records and included demographic data, referral route (ED, primary care or transfer from another hospital), intraoperative findings, procedures performed and follow-up outcomes.

In our unit, where testicular torsion is confirmed intraoperatively, testicular detorsion with bilateral three-point fixation (within a dartos pouch using a non-absorbable suture) is performed. Where the torted testis is deemed non-viable intraoperatively, orchidectomy with contralateral three-point testicular fixation is performed. Patients with suspected torsion are not managed with attempts at manual detorsion or assessed with testicular Doppler ultrasonography in our unit. In cases where the testis is not torted, our routine procedure is to return the testis to the scrotum via a dartos pouch. (Three-point fixation is not performed in all instances according to consultant preference.)

Data were collated and analysed with Excel^® (Microsoft, Redmond, WA, US). Statistical tests were performed with online software (VassarStats.net). For age and duration of symptoms at presentation, the Kruskal–Wallis test was employed to compare medians between three groups and the Mann–Whitney U test between two groups. For comparisons of orchidectomy rate between groups, the chi-squared test was used.

Results

Over the 8-year study period, 662 boys underwent emergency scrotal exploration. There were 6 internal referrals (1%), 294 (44%) attended our ED directly, 271 (41%) were referred from primary care (general practitioner [GP] or walk-in centre) and 91 (14%) were transferred from other hospitals. There were 105 children (16%) under 5 years of age. These comprised 4 of the internal referrals (66%, including 3 neonates), 36 from the ED (12%, 3 neonates), 33 from primary care (12%) and 32 transfers (36%, 8 neonates).

Exploration was performed within three hours of the decision being made to operate in 80% of the total cohort (528/662) but when considering only those in whom surgery was booked to be performed within three hours, the proportion rose to 91% (353/390). The median time from decision to operate to start of the procedure was 97 minutes (interquartile range [IQR]: 68–162 minutes).

All 14 neonates had intraoperatively confirmed testicular torsion. Orchidectomy with contralateral 3-point fixation was performed in 13 (93%) of these, with 1 (7%) undergoing testicular detorsion and bilateral 3-point fixation. There were no non-neonatal internal referrals with torsion. Subsequent analysis excludes neonates in view of the absolute requirement for specialist paediatric surgical care and the very high probability of a non-viable testis requiring orchidectomy being found at exploration.

Testicular torsion was present in 100 (15%) of the 648 non-neonates at the time of operation. Torsion of the hydatid of Morgagni was diagnosed in 335 (52%) and epididymo-orchitis in 86 (13%). There was an alternative diagnosis in 56 (9%), including hydrocoele, haematoma, epididymal cyst, cord oedema, idiopathic scrotal oedema, inguinal hernia and Henoch–Schönlein purpura. No abnormal findings were present at exploration in 51 (8%) of the non-neonates and there were incomplete records (unknown diagnosis) in 20 (3%). Six had a potential diagnosis of intermittent testicular torsion (intraoperative findings of cord oedema only [n=3] or lack of abnormal findings [n=3]) and were treated with bilateral three-point fixation within a dartos pouch. These cases are excluded from analysis of those with confirmed intraoperative diagnosis of testicular torsion.

Further analysis refers only to non-neonatal cases. Table 1 summarises the clinical details of the patients undergoing emergency scrotal exploration. Of the 100 boys with confirmed testicular torsion, 34 (34%) underwent orchidectomy, and 66 (66%) underwent testicular detorsion and bilateral fixation. The orchidectomy rate in the presence of torsion was significantly higher in transfers than in ED referrals (p=0.026, chi-squared test).

Table 1.

Source of referral and clinical details of boys with confirmed testicular torsion at emergency scrotal exploration. Neonatal cases and internal referrals are excluded.

	Emergency department	Primary care	Transfer from other hospital
Emergency scrotal explorations	291	271	83
Confirmed torsions (n=100)	52 (18%)	28 (10%)	20 (24%)
Orchidectomies (n=34)	12 (23%)	12 (43%)	10 (50%)
Median age in years	14 (IQR: 12–15)	13 (IQR: 9.3–14)	12.5 (IQR: 3.4–14.8)
Median symptom duration in hours	7.5 (IQR: 3.9–28.9)	14 (IQR: 6.4–24)	9.5 (IQR: 6.6–54)
Median time from arrival to operation in minutes	197 (IQR: 156–250)	199 (IQR: 144–268)	133 (IQR: 120–164)
Median time from booking to operation in minutes	88 (IQR: 64–115)	113 (IQR: 78–166)	80 (IQR: 57–112)

There was no statistical difference in median duration of symptoms at initial presentation to hospital (tertiary or local) (p=0.56, Kruskal–Wallis test) or median age (p=0.10, Kruskal–Wallis test) between referral groups. However, symptom duration at presentation in those who underwent orchidectomy (median: 30 hours, IQR: 12–72 hours) was significantly longer than in those who underwent testicular detorsion and bilateral fixation (median: 4.5 hours, IQR: 3.5–9 hours) (p<0.0001, Mann–Whitney U test). Of the 34 patients who underwent orchidectomy, 9 (24%) were under 5 years of age compared with 6 (9%) of the 66 patients who underwent detorsion and bilateral fixation (p=0.044, chi-squared test).

The median time from arrival in the tertiary unit to procedure start time was significantly shorter in the transfer group than in the ED and primary care groups (p=0.014, Kruskal–Wallis test) although there was no statistically significant difference in time from decision to operate to procedure start time (p=0.11, Kruskal–Wallis test). For boys who were transferred, the median delay between arrival at the local hospital and arrival in our unit was 167 minutes (IQR: 160–336 minutes) but data were only available for 6 of the 20 transferred patients.

Postoperative follow-up review was arranged for 49 (73%) of the 67 patients (including 1 neonate) who underwent testicular detorsion and bilateral fixation. Forty-four (66%) of these attended their appointment. The median time from surgery to initial follow-up appointment was 188 days (IQR: 114–400 days). The testis that was detorted at the time of exploration was of good size (symmetrical to the contralateral testis) in 38 (86%) of the 44 patients attending follow-up appointments and had atrophied in 6 (14%). Four of these six patients had originally presented to the ED, one was referred from a GP and one was transferred from another hospital. In the neonatal case, the testis was found to be of good size at seven years following surgery.

Discussion

Together with the British Association of Paediatric Surgeons and the British Association of Paediatric Urologists, The Royal College of Surgeons of England (RCS) published a commissioning guide accredited by the National Institute for Health and Care Excellence in 2016 for the management of paediatric torsion.¹⁰ The guide states:

Considering the time-critical nature […] of the condition, patients and their families will benefit from assessment and surgery performed locally. The transfer of a boy with a suspicion of torsion from a Secondary Care Institution to a Tertiary Centre should therefore be an exceptional occurrence (e.g. medical comorbidities).

In cases where appropriate management cannot be provided locally in a safe manner, urgent transfer to a tertiary centre is required because of the threat to organ survival.¹¹

The primary aim of this study was to determine whether boys undergoing emergency scrotal exploration with intraoperatively proven testicular torsion had a higher orchidectomy rate when transferred from other hospitals to our tertiary paediatric surgical unit than boys presenting directly to our ED or via primary care. The orchidectomy rate of 23% for the ED group was less than half the rate for those transferred from local hospitals (50%), a difference that was statistically significant. It is notable that outcomes were worse in the transfer group despite the median time from arrival in our unit to procedure start time being significantly shorter than for the other groups. Although the delay incurred by transfer could only be calculated in six cases, the implication is that interhospital transfer can delay definitive management by nearly three hours and this would be likely to contribute to a reduced prospect of testicular viability at operation.

The RCS commissioning guide suggests that exploration should always occur within three hours of the decision to operate.¹⁰ Although comparable with a recent report,¹² our rate of adherence to this recommendation (80%) falls short of this target, partly because the majority of patients were historically booked for surgery to be performed within six hours of the decision to operate. Recent practice in our unit has, however, involved booking patients for surgery to occur within three hours in order to adhere more closely to the commissioning guide. The testicular salvage rate of 86% at our median follow-up time of six months compares favourably with previously reported outcomes.¹³

Provision of a service to facilitate local management of paediatric torsion should be addressed by the collaboration of surgeons, anaesthetists and paediatricians in secondary care hospitals and tertiary paediatric surgical units within clinical networks to formulate agreements regarding the management of these boys, with a minimum age (in the absence of medical co-morbidities) above which they should be managed in the referring hospital. Higher paediatric anaesthetic training will have been delivered to all anaesthetists holding a Certificate of Completion of Training or equivalent in the UK, and as such, they should be competent to provide perioperative care for emergency surgical conditions in children aged three years and older.¹⁴

Support and training in the management of testicular torsion should be provided to adult surgeons and urologists as well as their trainees. Paediatric urology training forms part of the curriculum for higher urology surgical training in the UK, with scrotal exploration for testicular torsion being a specific competency.¹⁵ Adult general surgical trainees are also required to be competent in the management of acute testicular torsion in childhood¹⁶ and reducing the number of boys transferred to tertiary paediatric surgical centres should provide increased opportunities for adult trainees to achieve their indicative numbers for this procedure.

Fourteen per cent of the explorations performed in our unit over the study period were for transferred cases, equating to nearly one transfer patient a month. Two-thirds of these patients were boys aged ≥5 years. By introducing a system in which boys who could be explored locally are not transferred, the burden of these transfers on the ambulance service and on tertiary unit emergency theatres could be considerably reduced.

The East Midlands Clinical Network (of which our unit is the tertiary paediatric surgical centre) facilitated the RCS commissioning guide on paediatric torsion.¹⁰ An agreement was reached in 2016 whereby the secondary care institutions in the network undertook to manage paediatric torsion in healthy boys aged ≥5 years on a 24/7 basis. Transfer of boys to our tertiary unit who met the criteria for local management would only be considered following consultant surgeon review at the presenting hospital and referral based on unusual or exceptional circumstances would be considered on a case-by-case basis.

Prior to the implementation of this service, a training day was provided by paediatric surgeons from our unit for adult surgeons from the network hospitals. This included both lectures and clinical skills practice sessions on the management of paediatric torsion. The service was put into practice in our network from 1 January 2017, and a prospective study is in progress to determine its impact in both our unit and the local network.

Previous studies have reported associations between transfer of patients to tertiary units and increased risk of orchidectomy,^17,18 and adherence to the RCS recommendations should reduce the delay in definitive management. Reducing the number of urgent transfers for suspected testicular torsion also minimises the burden on emergency ambulance services in having to undertake these journeys (which entail additional risks to patient, ambulance crew and general public) along with the non-availability of the ambulance for other emergencies at a time of increasing demand.

A review of urological litigation in the National Health Service identified patients with testicular torsion as the single largest group (excluding cancer patents) in which failure to diagnose and treat the condition resulted in a successful claim.¹⁹ A claim of clinical negligence may arise should an adult surgeon transfer a child with suspected torsion to a tertiary unit where an orchidectomy is subsequently necessary owing to a non-viable torted testis. The RCS guidelines suggest that in the absence of mitigating circumstances (baby/infant or significant co-morbidities), the adult surgeon owes the child a duty of care.¹⁰ Electing to not operate locally could be considered a breach of that duty of care, with the delay in definitive management incurred by transfer being a potential causal factor in the injury sustained (orchidectomy), consideration being given to the duration of symptoms at initial presentation and therefore the potential for testicular viability.

Study limitations

Given its retrospective nature, there are some limitations to this study. These include the incomplete availability of patient data regarding time of arrival at the presenting hospital and time of onset of symptoms. As only boys who underwent surgical exploration were considered in this study, no analysis has been made of boys presenting or referred to our unit with acute scrotal pain who were subsequently not operated on (eg those with an obvious diagnosis of idiopathic scrotal oedema); the burden of interhospital transfers is therefore likely to have been underestimated by this study.

Conclusions

The results of this study provide evidence to support the RCS recommendation for undertaking paediatric scrotal explorations in the presenting hospital whenever safe and appropriate, rather than delaying definitive management and threatening testicular viability by transferring boys with an acute scrotum to a tertiary paediatric surgical unit.