Unintended overexposure of a patient during radiotherapy treatment at the Edinburgh Cancer Centre, in September 2015

Executive Summary

Introduction

Between 14 ^th September 2015 and 18 ^th September 2015, a patient received a dose of ionising radiation much greater than that intended while undergoing a course of radiotherapy at the Edinburgh Cancer Centre ( ECC) in Scotland. Since the incident resulted from a procedural error, rather than from equipment failure, it has been reported and investigated under the provisions of Statutory Instrument 2000 No. 1059, The Ionising Radiation (Medical Exposure) Regulations 2000 (as amended) [1] (referred to in this report as the IR( ME) Regulations). The regulator for the IR( ME) Regulations (the 'appropriate authority') in Scotland is the Scottish Ministers.

This report, by the Inspector warranted by the Scottish Ministers for the IR( ME) Regulations, records the findings of the incident investigation. It identifies the errors that caused this overexposure and includes consideration of the deficiencies that contributed to the errors and where responsibilities for these deficiencies lay. It also makes recommendations intended to minimize the possibility of recurrence of any similar errors and to enhance patient safety in radiotherapy more generally.

The nature and consequences the error

Between 14 ^th and 18 ^th September 2015, a patient diagnosed with multiple myeloma was given palliative radiotherapy treatment at the ECC. This involved irradiation of the vertebrae of the neck to address pain and disability being caused by a bone fragment from complete collapse of the third cervical ('C3') vertebra.

The treatment prescribed by the oncologist was a total dose of 20 Grays of X-ray radiation to be delivered in 5 fractions, each of 400 centiGrays (hundredth of a Gray normally abbreviated to 'cGy'). Each of the 400cGy fractions was to be divided into two 200cGy beams, one to be delivered from the left side of the neck, and one from the right. The oncologist wrote this information clearly and correctly in the patient's 'Radiotherapy Prescription Sheet'.

The prescribed radiation dose, method of delivery, and fractionation were as expected for treatment of this condition, and in accordance with the relevant ECC Employer's Written Protocol.

This method of treatment is known as 'lateral parallel opposed fields at 100cm FSD' (focus to skin distance). (The other the form of treatment used at the ECC for the 'cervical spine' is a single (posterior) beam from the back of the neck.)

A Radiographer trained in treatment planning (referred to in this report as Radiographer A) used the information in the Radiotherapy Prescription Sheet to calculate manually (as opposed to using a computer) the dose of radiation that should be delivered to each side of the neck to achieve the prescribed dose to the C3 vertebra. This involves consideration of the attenuation of the beam of radiation as it passes through the skin and underlying tissue. This is generally referred to as the 'depth-dose' calculation.

Radiographer A used the correct ECC treatment planning protocol, but made an error in the depth-dose calculation to the extent that the calculated doses to each side of the neck entered in the 'Radiotherapy Prescription Sheet' were 100% too high.

In accordance with ECC Protocols, a second treatment planner (Radiographer B), carried out a similar manual calculation, but made the same mistake, and got the same wrong answer.

The next step in checking the treatment plan was for patient information, including the result of the manual calculations, to be input to an electronic information management system for radiotherapy called Aria, from where the appropriate information is transferred electronically to a dose calculation programme called RadCalc.

Data input to Aria was carried out correctly by Radiographer C, and this included the (wrong) dose from the manual calculation. RadCalc then calculated the daily dose to each side of the neck independently, and flagged correctly that the manually calculated dose was incorrect.

A fourth Radiographer (Radiographer D) then tried to reconcile the RadCalc calculation with the result of the manual calculation but failed to do so. In accordance with ECC practice, the Radiographers involved therefore sought the assistance of a member of the ECC's Treatment Planning Section (Physicist A). The erroneous belief of the Radiographers at this stage was that the manual calculations were correct, and the computed result was in error. The manual calculation was not rechecked.

For reasons that have not been fully resolved during this investigation, Physicist A achieved an answer from RadCalc that appeared to agree, to within the defined tolerance of 2.5%, with the (erroneous) manual calculations of the number of 'monitor units' ( MU) to be set on the treatment machine (the 'Linac') for each treatment.

As a result of these errors the Linac was set to deliver a total of 40 Grays of radiation to the treatment area in 5 fractions of 800cGy, instead of the prescribed 20 Grays in 5 fractions of 400cGy.

The magnitude of this overexposure is such that there was a significant possibility of serious harm to the patient.

The error was first identified at the 'summary/finishing off' stage on 29 ^th September 2015, 11 days after completion of treatment.

Upon discovery, the overexposure was reported promptly by the ECC to the Inspector Warranted by the Scottish Ministers for the IR( ME) Regulations, and this initial notification was followed by a written incident report.

The circumstances of the error

Accurate treatment planning in radiotherapy is of critical importance for patient safety, and must be carried out by staff who are appropriately trained, using the correct written procedures, and with appropriate checks and verification of data to ensure that any errors that might arise are clearly identified and corrected. With regard to these requirements, the concerns identified by this investigation include the following.

The approach that is recommended and generally adopted in treatment planning is for the first manual calculation to be checked independently using a different method. In this case, however, two Radiographers used the same method of manual calculation of the MU for this patient and got the same wrong answer.

The method defined in the ECC Employer's Written Procedure for the use of RadCalc for parallel opposed pair treatments, while capable of producing the correct result, was not that which would recommended by the producers of this software. The description of this method in this Employer's Written Procedure had been changed in February of 2015, but with no retraining of the operators involved.

The Radiographers involved lacked understanding of the defined method of RadCalc usage, and had little confidence in the results arising. They therefore persisted in an assumption, that their erroneous manual calculations were correct and that RadCalc was in error.

Entitlement of radiographers for treatment planning did not include a separate requirement for written evidence of satisfactory completion of initial training in the use or RadCalc, or of any formal provision for maintaining their competence in its use.

Responsibilities

The findings of this investigation are that, in general, the duties of the employer under the IR( ME) Regulations were being implemented appropriately, but with some concerns regarding the provision and recording of operator training. None of the staff involved were found to have acted negligently or to have contravened any requirement of the relevant Employer's Written Procedures. It is also accepted that these operators genuinely believed that they were properly trained and experienced for the duties involved, although the nature of the errors made would suggest otherwise.

Recommendations and actions arising.

As a result of this incident the ECC has made a number of changes details of which are discussed further in the report.

Additional recommendations arising from this investigation include:

• A review of the current ECC procedures for manual calculations.
• A change in the current method of manual calculation to ensure that the second calculation uses a different method from the first.
• A review of the current ECC procedures for the use of RadCalc.
• Retraining of all operators in the use of RadCalc to ensure that they have an appropriate level of understanding of its workings and of confidence in its results.
• A review of the provision, maintenance, and recording of operator training relevant to this incident, and of the relationship between training and entitlement of duty holders.

Serious consideration has been given to the need for an Improvement Notice with regard to the provision and recording of operator training. However, given undertakings by the ECC that these deficiencies are already under review, such enforcement action has been deferred pending consideration of the outcome of this internal review and response to the recommendations of this report.

Unintended overexposure of a patient during palliative radiotherapy treatment at the Edinburgh Cancer Centre, in September 2015.