Neurosurgery clinical outcomes

Clinical outcomes are broadly agreed, measurable changes in health or quality of life that result from our care. Constant review of our clinical outcomes establishes standards against which to continuously improve all aspects of our practice.

About the Neurosurgery Service

The neurosurgical unit of Great Ormond Street Hospital (GOSH) is the largest paediatric neurosurgery unit in the United Kingdom, and the only paediatric neurosurgery unit in the North Thames area. The unit performs around 1000 procedures each year, providing neurosurgical input to various highly specialised multidisciplinary teams for the management of conditions such as:

As GOSH is one of only four supra-regionally funded craniofacial centres and one of the UK's four specialised Children’s Epilepsy Surgery Service (CESS) centres, the neurosurgery unit is responsible for an increasing volume of craniofacial and, in particular, epilepsy cases.

The following table shows the number of neurosurgical operations at GOSH for the past three years. Vascular activities do not contain intervention radiology for the last two quarters:

Number of neurosurgical operations

Type of operation Number of operations
2013/14 2014/15 2015/16
Brain Tumour 174 171 157
Cerebrospinal Fluid (CSF) related 153 181 196
Epilepsy 93 94 100
Craniofacial 163 203 198
Cranio-Cervical Junction 37 35 43
Cranio - Other 108 80 117
Spinal Tumour 8 14 10
Spinal Dysraphism 84 95 83
Spinal - Other 29 65 52
Vascular related 48 81 59
Total 926 1,084 1,067

Clinical outcome measures


1. Risk-adjusted mortality

2. Surgical adverse events

3. Shunt infection rates

4. Early shunt re-operation rates

1. Risk-adjusted mortality rates

The Neurosurgical National Audit Programme (NNAP) was established by the Society of British Neurological Surgeons in 2013 as part of a major quality improvement initiative. The programme aims to support neurosurgical units in the UK and Ireland to improve patient care, outcomes, safety, and experience by providing high quality, robust audit data that is analysed and presented in a consistent and clinically relevant way. Results are available for both adult and paediatric centres.

Fig 1.1: Risk-adjusted 30-day mortality rate, 2013 to 2016

Risk-adjusted 30-day mortality rates Neurosurgery Fig 1.1 2013-2016

© 2014-2017 University Hospitals Birmingham NHS Foundation Trust, all rights reserved

This ‘funnel plot’ diagram displays the mortality ratio for elective procedures in paediatric neurosurgery centres for the three years April 2013 to March 2016. The results are ‘risk-adjusted’, which means that the mortality rate of hospitals where operations are performed on higher risk patients is adjusted to account for these factors. This enables calculation of a predicted mortality rate, against which centres are benchmarked. Expected mortality for each centre will vary from chart to chart depending on the number of procedures performed and the risk profile of the patients treated in that time period. The horizontal yellow line represents the expected ratio. The solid black lines are known as the ‘control limits’. Risk-adjusted rates appearing between the control limits are within the normal range. The results of all centres for this period are within normal range. GOSH has more expected deaths than other centres because of patient risk factors and higher volume.

2. Surgical adverse events

The GOSH neurosurgical unit has a zero tolerance approach to adverse events including surgical complications. Every such event is reviewed in a weekly meeting attended by all members of the unit, and actions designed to prevent repetition are discussed. Adverse events are defined here as “Any untoward event related to a child’s admission to the neurosurgical unit that had the potential to increase their stay in hospital and/or produce a temporary or permanent worsening of their health.”

It is important to emphasise that while a number of reports use surgical complications as a measure of performance, within the neurosurgical department at GOSH we choose to use the adverse event rate (which may or may not be related to surgical complications) as we feel this better reflects the patient’s experience and provides a more transparent view of our performance. When things do not go according to expectation, it does not necessarily mean that an error has occurred, but nonetheless we do need to report and look into such events in order to foster a culture of openness. For example, if a child were to develop a chest infection following an operation, this is unlikely to be reported as a surgical complication but it would be reported as an adverse event.

Adverse events are graded as follows:

  1. No increase in hospital stay* and no neurological injury (threatened or actual)

  2. Increase in hospital stay* but no neurological injury (threatened or actual)

  3. An event that either threatened or caused neurological injury

  4. Death

*Includes readmission and/or extra procedure(s) with general anaesthetic 

Table 2.1: Surgical adverse events*

Year  Total Operations Grade 1 adverse events Grade 2 adverse events Grade 3 adverse events Grade 4 adverse events Percentage of adverse events Grades 3 and 4
2015/16 1,015 35 91 14 2 1.6%
2014/15 1,019 32 85 7 6 1.3%
2013/14 897 16 53 19 5 2.7%

*non-primary neurosurgical procedures such as muscle biopsies etc have not been included

Figure 2.2: Surgical adverse events by specialty*

Neurosurgery Surgical adverse events by specialty, 2013-14 to 2015-16

*non-primary neurosurgical procedures such as muscle biopsies etc have not been included

Table 2.2 Surgical adverse events by specialty*

Specialty Year Total operations Grade 1 adverse events Grade 2 adverse events Grade 3 adverse events Grade 4 adverse events

Percentage of adverse events Grades 3 & 4

Brain Tumour Surgeries 2015/16 157 2 14 4 0 2.5%
2014/15 171 3 21 4 2 3.5%
2013/14 174 3 16 6 0 3.4%
Cerebrospinal Fluid (CSF) related 2015/16 196 8 31 4 0 2.0%
2014/15 203 6 21 1 1 0.9%
2013/14 163 3 5 8 0 4.9%
Epilepsy Surgeries 2015/16 100 4 3 1 0 1.0%
2014/15 94 1 5 0 0 0.0%
2013/14 93 3 5 2 0 2.2%
Craniofacial Surgeries 2015/16 198 11 10 0 0 0.0%
2014/15 181 16 11 0 0 0.0%
2013/14 153 3 9 2 0 1.3%
Spinal Surgeries 2015/16 188 4 23 2 0 1.1%
2014/15 144 0 15 1 0 0.7%
2013/14 129 2 6 0 0 0.0%
Vascular related 2015/16 59 1 5 1 1 3.4%
2014/15 81 0 6 1 3 4.9%
2013/14 48 0 3 1 4 10.4%

*non-primary neurosurgical procedures such as muscle biopsies etc have not been included

Shunt operations

A shunt is a device that diverts accumulated cerebrospinal fluid (CSF) around obstructed pathways and returns it to the bloodstream. It is inserted in a neurosurgical procedure so that the upper end is in a ventricle of the brain and the lower end leads either into the heart (ventriculo-atrial) or into the abdomen (ventriculo-peritoneal). Permanent shunts are intended to stay in place for life. However, sometimes additional operations are required as a result of mechanical failure of the shunt, infection of the CSF, shunt blockage, or other reasons.

3. Shunt infection rates

Shunt infections are defined as: CSF microscopy/culture demonstrated an organism or there was CSF pleocytosis associated with fever, shunt malfunction or neurological symptoms, requiring shunt removal and subsequent antimicrobial treatment.

Table 3.1: Number and rate of shunt infections**

Year Total number of permanent shunt operations Number of shunt infections Percentage of shunt infections
2015/16 169 6 3.5%
2014/15 170 7 4.1%
2013/14 157 6 3.8%

**We have included all permanent shunt procedures, where the shunt was inserted, revised or internalised. We have included lumbar-peritoneal shunts, valveless shunts and subgaleal shunts, but excluded external ventricular drainage (EVDs).

4. Early shunt re-operation rates

Early shunt re-operation rate is defined as: Of all patients undergoing permanent shunt operations within a 12-month period, the number that required a second shunt operation with 30 days of the first operation for any reason.

Table 4.1: Number and rate of early shunt re-operations

Year Total patients for permanent shunt procedures Patients requiring re-operation within 30 days Percentage of re-operation
2015/16 120 21 17.5%


122 13 10.7%
2013/14 112 11 9.8%

A specific clinical reason for the increase in 2015/16 has not been identified. We continue to monitor the figures and discuss in clinical meetings to ensure that any change that is not natural variation is identified and acted upon.

This information was published in June 2017.