The impact of a managed clinical network on referral patterns of sarcoma patients in Grampian

Introduction

Bone and soft tissue sarcomas are malignancies of connective tissue, representing approximately 1% of all adult malignancies in the UK. ¹ These tumours are best dealt with referral to a centre specialising in their management, providing multidisciplinary care with orthopaedic surgeons, general surgeons, plastic surgeons, thoracic surgeons, oncologists, radiologists and pathologists. Chemotherapy and radiotherapy have improved the survival of sarcoma patients, while advances in limb salvage surgery have improved the quality of life of patients who would previously have required an amputation.

NHS Grampian serves a population of 530,000. It is a tertiary referral centre for bone and soft tissue sarcoma serving a population of approximately 1.2 million patients covering NHS Highland (population 310,000) and NHS Tayside (population 380,000). The report of the Scottish acute services review ² indicated the importance of managed clinical networks (MCNs) delivering a service with improved quality, access, convenience and coordination. ³ The incidence of sarcomas lends itself to the development of a MCN. These patients are best dealt with by specialist multidisciplinary teams (MDTs), which cannot be provided in every centre. The Scottish Sarcoma Managed Clinical Network (SSMCN) provides designated surgical and oncological centres thereby facilitating patient accessibility without compromising quality of care.

The SSMCN is based on the Scandinavian model. The development of the Scandinavian Sarcoma Group (SSG) in 1979 increased the number of multidisciplinary ‘tumour teams’, with regional responsibility for centralised treatment of sarcoma patients. The aim of the SSG is to uphold and improve the quality of diagnosis, treatment and care of sarcoma patients. ⁴

In 2004, the SSMCN was established with similar aims, to optimise and regulate the management of sarcoma patients by facilitating national multidisciplinary discussion of all bone and soft tissue sarcoma cases. As well as registering clinical, radiological, oncological and pathological details of all sarcomas presenting in Scotland ⁵ these cases are discussed at a weekly MDT teleconference meeting. Participating centres comprise MDT members from the three nationally designated surgical sarcoma centres in Aberdeen, Edinburgh and Glasgow as well as oncology and radiology services in Dundee and Inverness. In Grampian an orthopaedic surgeon specialising in surgical oncology undertakes all bone tumour resections the majority of soft tissue sarcomas excised from the trunk or extremity. When required, these procedures are performed with surgeons from other surgical specialities, e.g. plastic, general and cardiothoracic surgeons to ensure that the patients receive the best possible care.

It has been well documented that sarcoma patients benefit from treatment in a specialised sarcoma centre, with a Swedish study demonstrating a local recurrence rate three times higher in those patients treated outside a sarcoma centre. ⁶ Little evidence exists in the published literature as to whether managed clinical networking actually benefits patient management. We set out to assess what influence the SSMCN has had on referral patterns, appropriate investigations and surgical resection in Grampian.

Methods

A database of all Grampian sarcomas was generated using the Department of Pathology APEX SNOMED coding system. This was introduced in 1991 and a text search for the word ‘sarcoma’ in all histopathology reports from 1991 to 2009 was performed. All reports found were individually read to ensure validity. In order to minimise the effect of any coding errors, this was cross-referenced against a further database generated by SNOMED searching specific sarcoma subtypes. Only definite primary sarcoma diagnoses were included.

Tumours presenting outwith the trunk or extremity were excluded leaving a cohort of 302 patients. Grampian NHS policy is to destroy medical records after five years if patients have not been under review. This resulted in 144 notes (48%) being unavailable for review, the majority of which had presented prior to the initiation of the network.

A total of 158 patients' notes were reviewed retrospectively for information on referral, investigations, management and clearance patterns. If all referral data were not available on note review, general practices were contacted to establish the duration of symptoms.

Results

Case records of 158 patients (94 male, 64 females) were reviewed retrospectively, 79 patients before the establishment of the SSN in November 2004 and 79 after. Table 1 shows the patient demographics.

OPEN IN VIEWER

Table 1.

Demographics of patients.

	Before 2004	After 2004
Number of patients	79	79
Number female	33	31
Number male	46	48
Median age (Min, Max)	55 (9,82)	59 (15,92)

Referrals were received from a variety of sources as illustrated in Figure 1. Prior to November 2004 more patients were referred directly to the sarcoma service by GPs, while subsequently greater numbers presented from other hospital specialists. This referral pattern is also apparent in Figure 2, which demonstrates slightly more patients were seen by more specialities after 2004. Figure 3 also confirms the changing patterns in referrals in general and specifically those from primary care. Referral numbers peaked in 2005 and 2006 following the initiation of the network. OPEN IN VIEWER

OPEN IN VIEWER

Figure 2.

The total number of specialties seen by sarcoma patients.

OPEN IN VIEWER

Figure 3.

The total number of referrals and the number of referrals direct from general practice since 1992.

The time interval from receipt of a referral to initial assessment by the service improved from a median of 19.5 days to 10 days after the SSN was established (Table 2) (p = 0.016). However, during the same time period, the median interval between initial GP consultation and review by a sarcoma specialist increased from 35 to 41 days, although this trend does not reach statistical significance (p = 0.57). Table 3 demonstrates the greater number of specialities a patient sees prior to being seen by the sarcoma team, the longer it takes for definitive management to be undertaken.

OPEN IN VIEWER

Table 2.

The median total time of referral from GP to specialist surgeons and median time to being seen by the sarcoma service once a referral is received.

	Before 2004	After 2004	Statistical significance
Median total time (days) from GP to sarcoma service	35	41	p = 0.57
Median time (days) from referral to sarcoma service to being seen	19.5	10	p = 0.016

OPEN IN VIEWER

Table 3.

The more specialities seen by sarcoma patients increase the total referral time to definitive surgical management (p < 0.001).

No. of specialities seen	Median time of referral (days)
1	15.5
2	69.0
3+	92.5

Table 4 demonstrates that the number of patients undergoing investigation with a magnetic resonance imaging (MRI) scan prior to excision of the sarcoma, increased from 67% to 86% after the establishment of the network. This figure reaches statistical significance (p = 0.009). The biopsy groups were combined into appropriate biopsy (i.e. core, incisional and planned excisional biopsy) and inappropriate biopsy (i.e. unplanned incisional and ‘whoops procedures’ (the so called planned but inappropriate excision). There was an increase in the number of patients undergoing appropriate biopsy from 57% to 79% (p = 0.006). Data were available on the adequacy of surgical margins in 69 patients in each group. We looked at the resection margins, grouping them into complete and incomplete margins. Prior to 2004, 33 (48%) patients had documented complete resection and 36 (52%) were documented as incomplete. Since the establishment of the SSMCN this has increased to 56 (81%) complete margins and 13 (19%) incomplete with a statistical significance of <0.001. Both before and after the initiation of the SSMCN a group of 10 patients exist who only underwent biopsy of their sarcoma in Grampian. This group consists of those patients who were not considered fit for surgery or who had advanced metastatic disease precluding resection of their tumour. Also within this group are those who underwent biopsy locally but the extent of their surgical resection required that to be undertaken elsewhere.

OPEN IN VIEWER

Table 4.

The percentage of patients undergoing MRI scanning and appropriate biopsy prior to resection of sarcoma.

	Before 2004	After 2004	Significance
MRI	53 (67%)	68 (86%)	p = 0.009
Appropriate biopsy	45 (57%)	62 (79%)	p = 0.006

Discussion

Our results confirm a reduction in the time interval between a referral to the sarcoma service being received and the patient being seen. Our aim is to highlight that delay in appropriate referral of bone and soft tissue sarcomas results in delay to diagnosis and potentially suboptimal management by clinicians untrained in sarcoma surgery. If all patients with suspicious lesions were referred via the sarcoma network directly from general practice this would decrease the number of incomplete excisions carried out without the benefit of appropriate preoperative planning in the form of MRI scanning and biopsy. Perhaps the most likely explanation for this finding is an increase in the number of patients who were referred to the sarcoma service via a circuitous route involving several specialities. Specialities not regularly managing sarcoma patients often have no urgent access clinics available for suspicious lumps, further delaying review of these patients.

This study is not without its limitations, the main one being its retrospective nature. This in addition to the loss of medical records after five years results in deficiencies within our dataset. Patients presenting earlier, before the start of the SSMCN are the ones with the most information missing and hence those most affected. We are also limited by the regional nature of the audit, we specifically wanted to assess the local effect of the MCN and if the changes implemented had improved patient care. In Grampian, we are fortunate to have access to a searchable pathology database for all sarcomas to aid identification of patients. This is not available in all regions. Since the initiation of the sarcoma network we have developed local databases for sarcoma referrals and biopsy proven sarcomas.

The rarity of these tumours undoubtedly makes the initial diagnosis in general practice difficult. A GP may see only one soft tissue sarcoma in 20 years of practice and subsequently we have to raise awareness of worrying symptoms and signs. ⁷

Health promotion is an integral part of a MCN. GPs, as the ‘gatekeepers of the NHS’. are generally the first to see and potentially diagnose sarcomas. In order to ensure prompt and appropriate referral to sarcomas centres the SSMCN developed posters (Figure 4a and b) providing information regarding symptoms and signs that could represent both soft tissue and bone sarcomas. These were designed to educate GPs throughout Scotland and were sent to each practice with appropriate contact details relevant to their own region. It is undoubtedly time to re-emphasise these points. Our data would suggest that hospital specialists have become more informed with regards to appropriate management of sarcomas, with the sarcoma service receiving more referrals from this group. It would appear that GPs have not taken on board the message from the poster campaign. We aim to re-target this group with more posters, seminars and lectures, particularly focusing on those with limited musculoskeletal interest. OPEN IN VIEWER

Grimer et al. ⁸ published new guidelines for the management of soft tissue sarcoma following a consensus meeting of UK sarcoma specialists. Guidelines previously published by the European Society of Medical Oncology (ESMO) ⁹ and the National Comprehensive Cancer Network (NCCN) ¹⁰ were used as a basis but adapted to accommodate UK clinical practice. They recommend ‘any patient with a soft tissue mass that is increasing in size, has a size >5 cm or is deep to the deep fascia, whether or not it is painful, should be referred to a diagnostic centre with a suspected STS.’ Grimer ¹¹ has used a golf ball as a visual aid to guide diagnosis of a ‘worrying lump’. This should not only be directed towards primary care but also highlighted to our hospital based colleagues.

Conclusion

Our results demonstrate that within Grampian the SSMCN has had a positive effect on the referral patterns to the sarcoma service; we are diagnosing, investigating and managing more of the soft tissue and bone sarcomas presenting in Grampian than previously. We have also seen an improvement in the number of appropriate biopsies, MRIs and complete resection margins in our patients since 2004. In order to continue to improve management of these patients undoubtedly the service will experience an increase in referrals both from primary and secondary care. It is however impossible to increase the number of malignant tumours seen without an associated increase in benign lesions, services have to be established to accommodate this.