Differences in outcome between patients readmitted to index vs non-index hospital trusts after colorectal resection

The clinical consequences of readmission following major surgery in the English National Health Service are unknown. This study aimed to determine differences in outcome between patients readmitted to index vs non-index trusts after major surgery.

Adult patients who underwent colorectal resection in England in April 2006 to March 2017 were identified in the national Hospital Episodes Statistics dataset. Patients were included if they were readmitted as emergencies within 30 days of initial discharge. The primary outcome measure was all-cause mortality within 90 days of readmission. Comparisons between patients readmitted to index vs non-index trusts were adjusted for confounders using multivariable logistic regression. Rectal resection patients were a planned subgroup.

The readmission rate following colorectal resection was 15.1% (54 680/364 481), with 7.1% (3905/54 680) readmitted to a non-index trust. The 90-day mortality following readmission was 7.1% (3874/54 680) overall and 3.9% (652/16 736) in the rectal resection subgroup. The reoperation rate was 19.2% (10 498/54 680) overall and 23.1% (3859/16 736) after rectal resection. Mortality was significantly higher in non-index [10.9% (427/3905)] vs index trusts [6.8% (3447/507 75), adjusted OR 1.50, 95% CI 1.34-1.68, P < 0.001]. There was an annual average of 14.7 excess deaths in non-index trusts; only 1.9 of these followed surgical reoperation. In patients who underwent rectal resection, only 0.3 of the total 1.9 excess deaths each year in non-index trusts followed surgical reoperation.

Despite a statistical difference, the absolute number of excess deaths attributable to readmission to a non-index trust is very low, particularly amongst patients requiring reoperation.