A prospective cohort study characterising patients declined emergency laparotomy: survival in the 'NoLap' population.

Patients eligible for emergency laparotomy who do not proceed to surgery are not as well characterised as patients who do proceed to surgery. We studied patients eligible for laparotomy, as defined by National Emergency Laparotomy Audit criteria, from August 2015 to October 2016. We analysed the association of individual variables with survival and two composite scores: P-POSSUM and a general survival model. Out of 314 patients, 214 (68%) underwent laparotomy and 100 (32%) did not. Median (IQR [range]) follow-up was 1.3 (0.1-1.8 [0.0-2.5]) years for the cohort, 1.5 (1.1-2.0 [0.0-2.6]) years after laparotomy and 0.0 (0.0-1.1 [0.0-2.2]) years without laparotomy. There were 126/314 (40%) deaths in the follow-up period, 52/214 (24%) deaths after laparotomy and 74/100 (74%) deaths without surgery. Ninety out of 126 deaths (71%) were within one month of hospital admission. Patient variables were different for the two groups, which when combined in the general survival model generated background median (IQR [range]) life expectancies of 12 (6-21 [0-49]) and 4 (2-6 [0-36]) years, respectively, p < 0.0001. 'Poor fitness' precluded laparotomy in 74/100 (74%) patients. The decision to not operate involved a consultant less often than the decision to operate: 66/100 (66%) vs. 178/214 (83%), p = 0.001. Our study supports the contention that survival beyond 30 postoperative days could be predicted reasonably accurately. Survival in patients who did not have laparotomy was shorter than expected. Emergency laparotomy might have prolonged survival in some patients.