Draft dataset for initial audit of emergency laparotomy patients

This is a draft dataset for discussion. Specific points to note:

1.  The final dataset will include definitions to ensure consistency across sites. These are under development

2.  We propose to start collecting limited data on a wide range of operations, and refine this as necessary. There is a huge amount of data that could be collected. However at present the real need is to simply define the size and extent of the problem. The results of this will guide further data collection. We also feel it is important to include as many people as possible at this stage. A very detailed dataset is likely to need considerably more enthusiasm in order to obtain complete data.

3.  This is intended to be a minimum dataset, in order to broadly establish a baseline amongst as many hospitals as possible. You may well need to supplement it with additional data in order to address local issues. If everyone can include this as a minimum, then it will help greatly by collecting consistent data across sites.

4.  The current dataset can be completed in theatre, with limited follow-up at a later date to capture discharge/mortality. This should facilitate a higher completion rate.

5.  Additional work is likely to require collection of pre-op Early Warning Scores (eg on admission and highest in the 24hrs pre-op) and POSSUM data. This will allow some analysis of acute pre-op morbidity and standardisation of mortality rates. It will require a significant increase in time taken for data collection. If individual units do wish to collect it, we would be interested in supporting this, and using your experiences for subsequent audits.

Patient population

All patients >60 years old

All non-elective upper and lower-GI laparotomy, including return to theatre for any reason

Not including: straight forward appendicectomy

Gynaecological laparotomy (unless general surgeons involved)

Surgery related to organ transplant

Laparotomy following trauma/penetrating injuries

Data to collect

•  Unit demographics

Number of emergency laparotomies per year

Number of specialist colorectal surgeons

Dedicated colorectal on-call rota Y/N

Dedicated upper-GI on-call rota Y/N

•  Age at operation (rather than date of birth, which is not sufficiently anonymous)

•  Sex

•  ASA

•  This can be subjective and needs defining, but would serve as a simple measure of “risk” for the first phase1. Cook defines ASA II as “mild to moderate systemic disturbance caused either by the condition to be treated surgically or by other pathophysiological processes”2. We need consistency to avoid some observers including only conditions which existed (chronically) before the acute episode in their ASA score, whereas others will include both chronic and acute changes. We also need consistency with regard to eg >80yo is ASA 2.

•  Date of admission

•  Admitting speciality

•  Existing local audit has identified that about 20% of patients were admitted to non-surgical specialities in the 1st instance. Part of the reasons for poor outcome may be because they spend too long under non-surgical specialities. It also tells us who we need to engage, ie physicians as well. Subsequent work might be to highlight the pathway to surgery.

•  Time & date booked for theatre.

•  An attempt to get at delays/access to theatre. Adequate access to theatres has historically been an issue. At this stage we are not collecting reasons for delays.

•  NCEPOD classification:

•  The tradition classification used scheduled / urgent / emergency. This has been changed fairly recently and now consists of immediate / urgent / expedited / elective. Definitions are attached and these need to be used rather than the traditional classification.

•  Date of surgery

•  Time entered anaesthetic room

•  This is easier to define than when surgery started. Used in conjunction with time/date booked for theatre, it provides information about urgency and delays.

•  Grade of most senior surgeon present in theatre.

•  Specialist colorectal / Upper-GI surgeon (as relevant)

•  ie was upper-GI surgery carried out by an upper-GI surgeon, similarly for lower-GI surgery, as opposed to a surgeon with a different sub-speciality practice (eg, breast surgeon who is on the general surgical on-call rota)

•  Grade of most senior anaesthetist present in theatre.

•  Operation performed: Oversew of upper-GI perforation

Small bowel resection right/left/sigmoid/subtotal/total colectomy

Hartmans Procedure

Ant/AP resection

Defunctioning ileostomy/colostomy

Other

•  Underlying pathology

•  Clinical rather than pathological diagnosis (UC, Crohns, Diverticular mass, strangulated/incarcerated hernia, malignancy, perforation, ischaemic bowel, adhesions, other)

•  Required level of post-op care following surgery: Level 3/ Level 2 / ward

•  Actual level of post-op care following surgery: Level 3/ Level 2 / ward

•  Ventilated in theatre/recovery whilst awaiting creation of ICU bed

•  Date of discharge

•  rather than fitness to discharge. If there is delay due to social reasons etc then this is part of the problem that needs to be tackled

•  (Level 3 length of stay / Level 2 length of stay)

•  this is useful with regard to defining the impact these patients have on critical care resources, but it requires a considerable increase in effort to collect the data. Comments over the feasibility would be welcomed.

•  Alive at “discharge” Y/N

•  Alive at 30 days Y/N

•  These provide both 30 day mortality and in hospital mortality.

1.  Donati et al. A new and feasible model for predicting operative risk. BJA 2004; 93: 393-399.

2.  Cook et al. Hospital mortality after urgent and emergency laparotomy in patients aged 65 years and over. Risk and prediction of risk using multiple logistic regression analysis. BJA 1998; 80: 776-781.

3.  Copeland PM. The POSSUM system of surgical audit. Arch Surg 2002; 137. www.archsurg.com.