Evaluating the clinical and cost effectiveness of pre-operative group and screen samples and routine post-operative bloods for Robot Assisted Radical Prostatectomy (RARP)
Nathan A 1, 2, Hanna N2, Rashid A2, Cleaveland P1, Appukutty J2, Miah S2, Collins J1, Sridhar A1, Kelkar A1, Briggs T1, Kelly J1, Shah N2, Lamb B2, Shaw G1, Sooriakumaran P1, Rajan P1, Nathan S1

1.University College London Hospitals, United Kingdom
2. Cambridge University Hospitals, United Kingdom

Introduction
Prostate Cancer is the most common cancer in the UK for men (CancerResearchUK, 2019) and is the most commonly diagnosed cancer in England (PublicHealthEngland, 2020). Between April 2017 and March 2018, approximately 400,000 men were living with prostate cancer in the UK, with an incidence rate of 42,668. During this period, 13,891 men had radiotherapy and 7,018 men had radical prostatectomy. 85% of radical prostatectomies were carried out with robot assistance, 8% open and 6% laparoscopically (NPCA, 2019).
Robot assisted surgery has the advantages of smaller incisions, reduced pain, shorter length of stay and improved surgical ergonomics. Robot Assisted Radical Prostatectomy (RARP) surgery has better outcomes for patients compared to open surgery. RARP achieves lower blood loss, reduces operation-time, reduces length of stay (LOS) and has lower transfusion rates compared to open surgery (Ilic et al., 2018) (Haese et al., 2019).
Patients undergoing RARP commonly require two group and screen samples pre-operatively. This practice dates from the period when open prostatectomy was most commonly performed. Overall transfusion rate in 2018 from 9747 cases was 0.15%. Open surgery had a transfusion rate of 0.78%, laparoscopic technique was 0.49% and robot assisted surgery was 0.08% (BAUS, 2019). Furthermore, day one post-operative blood tests are routinely undertaken in urology centres around the world, in order to check blood count and renal function. This also dates from the era of open surgery, when blood loss was higher, and anastomotic urine leak more common (Ilic et al., 2018).

Aims
We aim to assess the need for two group and screen samples pre-operatively and the usefulness and need of routine post-operative blood tests. Furthermore, we aim to suggest specific guidelines for requesting post-operative blood tests, to maintain patient care and safety, reduce costs and reduce discharge delays.

Methods
Ethical approval was obtained from trust audit committees in line with local and national guidelines.
1039 patients were retrospectively reviewed from two large specialist Urology centres in the United Kingdom. Patients underwent either primary or salvage RARP between 2017 and 2019. Data collected included; patient demographics (Age, BMI, ASA, co-morbidities), oncology (Gleason, T-staging), intra-operative data (pelvic lymph node dissection (PLND), nerve sparing, op-time, estimated blood loss (EBL), transfusion, complications) and post-operative data (clinical signs and symptoms, LOS, transfusion, complications). Additionally, 54 patients were prospectively assessed with implementation of new blood test requesting guidelines.

Results
Pre-Operative
The median age was 64 (IQR: 59-68), median ASA 2, median BMI 28 (IQR: 25-30) and 27% had Charlson co-morbidities [Table 1.1].
Intra-Operative
36% underwent PLND, 41% had bilateral nerve sparing, 42% had unilateral nerve sparing and 17% had no nerve sparing. Median op-time was 142 mins [IQR:120-170] and median EBL was 200mls [IQR:100-300] [Table 1.1].
The intra-operative complication rate was 1% (N=15). 0.6% (6) had an anastomotic urine leak, 0.5% (5) bowel injury, 0.2% (2) vascular injury, 0.1% (1) bladder injury and 0.1% (1) port-site injury [Table 1.2]. 0.3% (3) patients had an intra-operative transfusion.
Post-Operative
16% (N=162) of patients had clinical signs and symptoms flagged overnight or day 1 post-op. 6% (59) increasing abdominal pain, 3.2% (33) high drain output or drain leak, 2% (25) cardiovascular dysfunction, 1.6% (17) nausea or vomiting, 1% (15) low urine output, 0.4% (4) wound concern, 0.4% (4) desaturations, 0.3% (3) bleeding in stool or urine and 0.1% (1) calf pain [Table 1.3].
The overall post-operative complication rate was 4% (N=42). 1.4% (15) haematoma, 0.7% (7) anastomotic urine leak, 0.5% (5) ileus, 0.4% (4) bowel injury, 0.4% (4) wound breakdown, 0.4% (4) AKI, 0.1% (1) renal injury, 0.1% (1) ureteric obstruction and 0.1% (1) a DVT [Table 1.4]. 0.6% (6) had a post-operative transfusion. Median LOS was 1 day (IQR: 1-2).
Analysis
Of the 1039 patients in our series, 3% went to higher dependency post-operatively. 1% (15) had intra-operative concerns of which 1% (11) had complications. 16% (161) had post-operative concerns of which 3% (30) had complications. 80% (833) did not have any clinical reasoning for blood tests and all had routine blood tests of which zero patients developed a complication [Figure 1.1]. 1 patient was re-admitted after developing a haematoma which was managed conservatively. He was discharged with normal clinical assessment and normal bloods (Hb128) post-operatively. The sensitivity of clinical assessment and routine blood tests in our series was 98%, specificity was 0% (all patients had routine bloods), positive predictive value was 4% and negative predictive value was 0%. Furthermore, only 43% of patients with a post-operative complication had abnormal day 1 blood results that were beneficial diagnostically.
Re-Audit Results
Guidelines developed from analysis of the initial audit data were used to specifically request post-operative blood tests [Figure 1.2]. Results were prospectively re-audited for 54 patients in one trust over a one-month period post implementation of new guidelines. Demographics were similar to the initial cohort of patients. Median age was 61 (IQR: 56-67). Median ASA was 2, 28% had Charlson co-morbidities and 28% had PLND. Median op time was 110 mins (IQR: 85-152.5) and median EBL was 300mls (IQR: 200-500). There were no intra-operative complications or blood transfusions. There was one post-operative complication (haematoma) and no post-operative blood transfusions. 11% (N=6) of patients had an intra-operative request for post-operative bloods [5: >500mls EBL, 1: salvage case]. 28% (N=15) of patients had post-operative clinical concerns that triggered a request for post-operative bloods [5: high drain output, 2: increasing abdominal pain, 4: cardiovascular dysfunction, 1: nausea, 1: oxygen requirements, 1: wound concern, 1: confusion]. In total 39% (N=21) of patients were flagged by the guidelines as concerning and needing post-operative blood tests. The one post-operative complication was successfully flagged by the guidelines and managed successfully.
Re-Audit Analysis
In total the number of bloods requested dropped from 100% to 39%. No post-operative complications were missed. Discharge delays reduced from 10% to 0% (in one specific trust). Overall sensitivity of the new guidelines in diagnosing post-operative complications improved to 100%, specificity improved to 62%, positive predictive value improved to 5% (due to the low prevalence of complications) and negative predictive value improved to 100%.

Discussion
In our series of 1039 patients, intra-operative and post-operative clinical assessment successfully highlighted complications in 99.9% (1038) of patients. Routine post-operative blood tests (80%), without an indication, did not flag any complications missed by clinical review. Furthermore, day 1 post-operative blood tests were only of diagnostic benefit in 43% of complications (18 / 42). Therefore, emphasis should be placed on clinical assessment rather than blood tests to diagnose post-operative complications.
The NHS is currently £14 billion in debt with increasing year-on-year liabilities (Dunhill, 2019). Additionally there are 'record numbers of NHS staff quitting' due to long hours and high workload (HealthEducationEngland, 2019). There is a strong local, regional and national objective to reduce expenditure and improve work satisfaction in the NHS. Unnecessary bloods tests waste clinician time, disturb patient rest, increase cost to the trust and result in discharge delays. Our cost-analysis approximates a minimum of £100 per RARP patient is being wasted due to unnecessary blood tests and discharge delays as a consequence of untaken, incorrectly processed, delayed or unactioned blood tests or results. Additionally, intra-operative (0.3%) and post-operative (0.6%) transfusion rate is low and requiring two group and screen samples is a further logistical and cost burden.
In 2018, 79,471 elective operations were cancelled in the NHS England due to non-clinical reasons with this number expected to rise for 2019 data (NHSEngland, 2020). 6% of our series had a discharge delay of one day or more due to delayed or incomplete blood tests. Reducing the number of surplus blood tests will improve patient turnover and increase the number of hospital beds available.
Future work should be aimed at assessing other surgical procedures and auditing the necessity of routine post-operative bloods across the NHS. Additionally, removing unnecessary investigations and reducing time-to-discharge could widen the adoption of outpatient RARP. Furthermore, a convolutional neural networking artificial algorithm could be used to flag patients at risk of complications and intervention could be started earlier.

Conclusion
In conclusion, routine post-operative blood tests are not necessary in RARP. Intra-operative and post-operative clinical judgement is highly sensitive in diagnosing complications. Specific guidelines for requesting post-operative blood tests can improve sensitivity and specificity of diagnosing a complication. Furthermore, the guidelines reduce the number of blood tests requested and reduce discharge delays to help the financial, logistical and resource burden on NHS trusts.

Recommendations
We have suggested an established and tested set of guidelines [Figure 1.2] to safely and accurately base requests for post-operative blood tests for RARP patients. The guidelines are based upon worldwide established literature of outcomes and complications and tailored to our audit of 1039 RARP patients. Careful consideration of indication can reduce blood test requests by 61% and eradicate blood test related discharge delays. We recommend these guidelines are adhered to for RARP and are mirrored with adjustments for all surgical procedures where the norm is to routinely request bloods. Additionally, blood transfusion rates for RARP are low, and trusts should review their group and screen request policy.

References:BAUS. 2019. Analyses of Radical Prostatectomies performed between January 1st and December 31st 2018 [Online]. BAUS. Available: https://www.baus.org.uk/_userfiles/pages/files/publications/audit/Prostatectomy2018finalanalyses.pdf [Accessed 30/01/2020 2020].
CANCERRESEARCHUK. 2019. Prostate Cancer Statistics [Online]. Available: https://www.cancerresearchuk.org/health-professional/cancer-statistics/statistics-by-cancer-type/prostate-cancer/incidence#ref-0 [Accessed 30/01/20 2020].
DUNHILL, L. 2019. Finance and Efficiency [Online]. Available: https://www.hsj.co.uk/finance-and-efficiency/nhs-trusts-owe-government-14bn/7025771.article [Accessed 25/02/20 2020].
HAESE, A., KNIPPER, S., ISBARN, H., HEINZER, H., TILKI, D., SALOMON, G., MICHL, U., STEUBER, T., BUDAUS, L., MAURER, T., TENNSTEDT, P., HULAND, H. & GRAEFEN, M. 2019. A comparative study of robot-assisted and open radical prostatectomy in 10 790 men treated by highly trained surgeons for both procedures. BJU Int, 123, 1031-1040.
HEALTHEDUCATIONENGLAND. 2019. Record numbers of NHS staff quitting due to long hours [Online]. Health Education England. Available: https://www.theguardian.com/society/2019/feb/17/nhs-staff-quit-record-numbers-work-life-balance [Accessed 25/02/20 2020].
ILIC, D., EVANS, S. M., ALLAN, C. A., JUNG, J. H., MURPHY, D. & FRYDENBERG, M. 2018. Laparoscopic and robot-assisted vs open radical prostatectomy for the treatment of localized prostate cancer: a Cochrane systematic review. BJU Int, 121, 845-853.
NHSENGLAND. 2020. Cancelled Elective Operations Data [Online]. NHSEngland. Available: https://www.england.nhs.uk/statistics/statistical-work-areas/cancelled-elective-operations/cancelled-ops-data/ [Accessed 25/02/20 2020].
NPCA. 2019. National Prostate Cancer Audit Annual Report [Online]. NPCA. Available: https://www.npca.org.uk/content/uploads/2020/01/NPCA-Annual-Report-2019_090120.pdf [Accessed 30/01/2020 2020].
PUBLICHEALTHENGLAND. 2020. Cancer registration statistics: England 2018 [Online]. PublicHealthEngland. Available: https://www.gov.uk/government/statistics/cancer-registration-statistics-england-2018 [Accessed 30/01/2020 2020].