Choosing Wisely Canada: Pediatric Neurosurgery Recommendations

Abstract

Objectives

Choosing Wisely Canada is an evidence-based, patient-focused, physician-led campaign to improve the delivery of medical care in Canada. The goal of this study was to produce Canadian recommendations for physicians treating patients with selected paediatric neurosurgery issues.

Methods

Paediatric neurosurgeons practicing in Canada were invited to participate. Suggestions were obtained using an anonymous questionnaire, and then ranked anonymously by the participating surgeons. Suggestions that received consensus from participants were discussed at the 2016 annual Canadian Pediatric Neurosurgery Study Group meeting. Suggestions that were not evidence based, or that would not have a substantive population impact were eliminated. All remaining suggestions were anonymously ranked by the group and the top five recommendations were submitted to Choosing Wisely Canada.

Results

The final five recommendations include: 1) don’t order a computed tomography scan to investigate macrocephaly (order an ultrasound or magnetic resonance imaging scan); 2) don’t image a midline dimple related to the coccyx in an asymptomatic infant or child; 3) don’t use computed tomography scans for routine imaging of children with hydrocephalus. Fast sequence nonsedated magnetic resonance imaging scans or ultrasounds provide adequate information to assess patients without exposing them to radiation or an anesthetic; 4) don’t recommend helmets for mild to severe positional flattening; 5) don’t do routine surveillance imaging for incidentally discovered Chiari I malformation.

Conclusions

Five Choosing Wisely Canada recommendations were produced to support care of patients with paediatric neurosurgical issues. While these recommendations will apply to the majority of children with the involved conditions, occasionally, deviation from these recommendations may be clinically indicated.

Choosing Wisely is an evidence-based, patient-focused, physician-led campaign that began in the USA in 2012, and in Canada in April 2014, to start a conversation about resource overutilization in medicine. The campaign’s goal is to ultimately reduce the utilization of unnecessary, and sometimes even harmful, tests and procedures, thus improving quality of care while reducing health care costs (1). Several medical and surgical societies have produced Choosing Wisely lists, including general neurosurgery (2–5); however, there are currently no Choosing Wisely guidelines published specifically for paediatric neurosurgery. The objective of this study was to produce a set of evidence-based, patient-focused, physician-led ‘Choosing Wisely Canada’ recommendations for paediatric neurosurgery, in order to reduce the utilization of unnecessary/harmful tests and procedures and thereby improve the quality of paediatric neurosurgery care in Canada while reducing costs. Note that Choosing Wisely recommendations are not meant to be rigid standard of care guidelines, though they should apply to most patients with the involved conditions.

Methods

Ethics approval was obtained for this study from our institutional research ethics board (CW15-0272/H15-02076). The members of the Canadian Pediatric Neurosurgery Study Group (CPNSG), a comprehensive body composed of the majority of neurosurgeons providing paediatric neurosurgery care in Canada, were contacted through e-mail addresses held by the CPNSG.

The purpose and methodology of the study were explained in an introductory e-mail. Each surgeon was asked to participate by completing a linked questionnaire. Two reminder e-mails, 2 weeks apart, were sent to maximize participation. This first questionnaire offered examples of established evidence-based Choosing Wisely recommendations for neurosurgery in the USA (5), and then asked for up to five suggestions for Canadian paediatric neurosurgery Choosing Wisely recommendations.

Following the two reminder e-mails, all the survey responses were compiled and a new questionnaire containing a list of all unique suggested recommendations was created and sent to each participant who completed the first questionnaire. This second questionnaire asked each participant to rank each suggestion from -2 to 2 (-2: they strongly feel it should not be a recommendation, -1: they weakly feel it should not be a recommendation, 0: they are equivocal about whether it should be a recommendation, 1: they weakly feel it should be a recommendation, 2: they strongly feel it should be a recommendation). Two reminder e-mails were sent, 2 weeks apart, to encourage participation and maximize the response rate.

The results of the second questionnaire were then analyzed and the suggested recommendations ranked from those with the highest number of points to those with the lowest. Suggestions with a total number of points ≤ 0 were rejected from further discussion. A list of suggestions with > 0 points were compiled in rank order from greatest number of points to the lowest number of points. This list was then presented to the CPNSG at the annual meeting in Whistler, BC in 2016. From this list, suggestions were eliminated if they were felt by the group to lack strong evidence, or if they did not apply to a broad primary care as well as a paediatric neurosurgery audience.

A final questionnaire was then sent out to the entire CPNSG group, asking each participant to rank each of the final recommendation suggestions from -2 to 2 as in the second questionnaire. The top five suggestions with the highest score were then selected to represent the Choosing Wisely Canada recommendations for paediatric neurosurgery and presented to Choosing Wisely Canada for approval and endorsement.

Results

A total of 35 paediatric neurosurgeons in Canada were invited to complete Questionnaire 1, and 12 neurosurgeons completed the questionnaire, producing 27 unique suggestions (Figure 1). These 12 neurosurgeons were subsequently invited to complete Questionnaire 2. Questionnaire 2 was completed by eight neurosurgeons, and seven suggestions were eliminated for having a sum score of ≤ 0. The remaining 20 suggestions were presented at the CPNSG 2015 annual meeting, and 11 suggestions were eliminated by the group. The remaining 9 suggestions were presented to all 35 paediatric neurosurgeons in Questionnaire 3, 15 neurosurgeons completed the questionnaire, and the top 5 highest ranked suggestions were chosen to be the final 5 recommendations for paediatric neurosurgery. These five recommendation suggestions were submitted to Choosing Wisely Canada for their assessment and endorsement (Table 1).

Figure 1.

The selection process for recommendations.

Table 1.

Choosing Wisely Canada: Pediatric neurosurgery recommendations

Don’t order a CT to initially investigate macrocephaly (order an ultrasound or MRI)
Don’t image a midline dimple related to the coccyx in an asymptomatic infant or child
Don’t use CT scans for routine imaging of children with hydrocephalus. Fast sequence nonsedated MRIs or ultrasounds provide adequate information to assess patients without exposing them to radiation or an anesthetic
Don’t recommend helmets for mild to severe positional flattening
Don’t do routine surveillance imaging for incidentally discovered Chiari I malformation

The final Choosing Wisely Canada pediatric neurosurgery recommendations.

Discussion

The final five recommendations and the evidence behind them

Don’t order a CT to initially investigate macrocephaly (order an ultrasound or MRI)

A common paediatric neurosurgery referral is a young child with a rapidly increasing head circumference crossing percentiles. The differential diagnosis is broad (6) and includes benign expansion of the subarachnoid spaces, subdural collections, hydrocephalus and neoplasm. When the fontanelle is open, the etiology can usually be diagnosed on head ultrasound, and this should therefore be the initial screening test of choice. In the absence of an open fontanelle, or if there are other signs and symptoms of acute raised intracranial pressure (vomiting, headache, irritability, drowsiness, full fontanelle, sun setting eyes), the etiology should be diagnosed with magnetic resonance imaging (MRI), if available, in order to limit radiation exposure. There is growing evidence that exposure to radiation through computed tomography (CT) imaging increases a child’s life long risk of cancer (7–10), and so all care should be taken to minimize this exposure as much as possible. Ultrasound (when fontanelle open), and/or MRI (when fontanelle closed), are therefore the screening tests of choice to investigate macrocephaly.

Don’t image a midline dimple related to the coccyx in an asymptomatic infant or child

Sacrococcygeal dimples (also called simple sacral dimples or sacrococcygeal pits) are common findings in newborns, with a prevalence of approximately 2% to 5% (11). They are not associated with any increased risk of occult spinal dysraphism (e.g., low lying conus, fatty filum, lipomyelomeningocele, split cord malformation, dermal sinus tract, etc.) compared with the general population of infants without sacrococcygeal dimples (11,12). There is therefore no need to investigate infants with this finding, with either ultrasound or MRI. Red flags for which investigating would be indicated include the presence of midline tuft of hair, sacral dimple or sinus tract above the gluteal cleft, hemangioma, dermal appendage and/or a subcutaneous lump (13). The ideal choice for initial investigation (ultrasound or MRI) would depend on the infant’s age, and the specific cutaneous findings and clinical symptoms present.

Don’t use CT scans for routine imaging of children with hydrocephalus. Fast sequence nonsedated MRIs or ultrasounds provide adequate information to assess patients without exposing them to radiation or an anesthetic.

Children with hydrocephalus, on average, obtain two head imaging assessments annually until the age of 20 (14). Their lifetime increase risk of fatal cancer is estimated to be 1 excess case of fatal cancer per 97 patients if standard head CT is used, or 1 excess case of fatal cancer per 230 patients if low-dose head CT is used (14). Head ultrasound (in infants with open fontanelles), and rapid sequence MRI do not require ionizing radiation and adequately assess for radiographic change in ventricle size (15–18). A rapid sequence MRI can be obtained without sedation and in under 3 minutes (18). It is therefore recommended that ultrasound (in infants with open fontanelles), or rapid sequence MRI (in all other children) be used for surveillance imaging in hydrocephalus at minimum, and ideally in emergency assessments as well when available. In the emergent setting, or when MRI is not available, low-dose noncontrast CT is appropriate.

Don’t recommend helmets for mild to severe positional flattening

Positional flattening is very common, affecting up to 40% of infants (19) since the Back to Sleep campaign began in 1992. There is now prospective, randomized control trial evidence that helmeting is no better at improving head shape in mild to severe positional flattening compared with physical therapy and providing general positioning recommendations such as maximizing tummy time while awake, and limiting time in swings and car seats (20). New guidelines from the Congress of Neurological Surgeons, following a systematic literature review including a review of the randomized trial mentioned above, consider helmeting as an option for severe cases of positional flattening (21). The prevalence of positional flattening in teens from the era following the Back to Sleep campaign but before helmets were widely used was less than 2%, suggesting that regardless of both the intervention used and the severity of the flattening, the vast majority of cases of positional flattening will cosmetically normalize (19). The cost of helmeting is also significant; a helmet costs thousands of dollars, and requires frequent adjustments over several months to adjust to an infant’s growing head (22). There are also risks associated with helmeting, including pressure sores and interference with parental attachment. With its associated high cost and only very weak evidence of benefit in treating positional flattening, there is no clear additional value in recommending helmets for infants with mild to severe positional flattening in addition to traditional positioning recommendations and physiotherapy.

Don’t do routine surveillance imaging for incidentally discovered Chiari I malformation

Chiari I malformation, defined as cerebellar tonsillar herniation greater than or equal to 5 mm below the foramen magnum on MRI brain, is a frequent incidental finding in children, with an estimated prevalence of 1% to 3% (23–25). The vast majority of children with incidentally discovered, asymptomatic Chiari I malformations have no clinically significant progression of tonsillar descent on routine follow-up, and symptom development is often unassociated with radiographic change (26,27). Radiographic follow-up in the absence of new symptomatology is therefore unnecessary.

Limitations

There is an inherent risk of selection bias from using questionnaires and group discussions at an annual meeting to develop these recommendations, as the response rate was less than 50%. Still, due diligence was taken to ensure that the vast majority of paediatric neurosurgeons practicing in Canada had the opportunity to participate in this study, many different centres and provinces were represented by participants, and consensus agreements from the CPNSG meeting ensured that there was strong general support for the recommendations.

Another potential limitation is that expert opinion was heavily relied upon in developing these recommendations, since strong support among paediatric neurosurgeons practicing in Canada was desired. Nonetheless, care was taken to ensure the final recommendations were rooted in evidence from peer reviewed published literature, so having supportive expert opinion should simply improve buy-in and strengthen the impact of these recommendations.

Finally, these recommendations should apply to the majority of children with the involved conditions, but they are not meant to be rigid standard of care guidelines. If unique circumstances apply in individual cases, health practitioners should use their clinical discretion to deviate from these recommendations as needed.

Conclusions

Choosing Wisely Canada is a campaign to improve the efficiency and quality of healthcare delivery in Canada. It provides recommendations, based on evidence regarding investigations and treatments that have been found to be ineffective and/or harmful. We have produced five evidence-based recommendations for paediatric neurosurgery, with the goal of improving the quality of paediatric neurosurgical care in Canada.

Ethics: Ethics approval was obtained for this study from the UBC C&W Research Ethics Board (CW15-0272/H15-02076).