Gysling, Savannah
(2024)
The effect of diabetes mellitus on outcomes following colorectal resections.
MRes thesis, University of Nottingham.
Abstract
Introduction
Diabetes mellitus affects over 9% of the global population, with an increasing prevalence over the last 15 years. Haemoglobin A1c (HbA1c) is a known measure of long-term diabetic control and can be used as a treatment target and referral threshold for patients with diabetes receiving surgery.
Diabetes is known to predict poorer outcomes in general medical and surgical patients, however the impact of diabetes on outcomes after colorectal resections has yet to be fully quantified. This thesis aimed to investigate the effect of diabetes and insulin-use on outcomes following colorectal resections using a national cohort and explore the perioperative use of HbA1c in current practice.
Methods
All adult patients with a recorded colorectal resection in England between 2010 and 2020 were identified from Hospital Episode Statistics (HES) data linked to the Clinical Practice Research Database (CPRD).
The primary outcome was 90-day mortality. Secondary outcomes included hospital length of stay (LOS), unplanned hospital readmissions within 90 days of discharge and the proportion patients with diabetes with HbA1c measured within one year before surgery.
Adjusted hazard ratios (HR) were calculated using Cox proportional hazard models for 90-day mortality and 90-day readmission. Length of stay was analysed using a loglogistic accelerated failure time model, with results presented as adjusted time ratios (TR).
Results
106,139 (50% male) patients were included with a median age of 66 (Interquartile Range (IQR) 52 to 76). The prevalence of diabetes was 10% (n = 10,931), encompassing 653 (6%) patients with type 1 and 10,278 (94%) patients with type 2 diabetes. 20% of all patients with diabetes had a record of insulin-use and 70% had a recorded HbA1c measured within one year before surgery.
The overall unadjusted 90-day mortality risk was 5.7%, with an increased hazard rate for patients with diabetes (adjusted HR 1.28, 95% CI 1.19 to 1.37, p < 0.001) compared to patients without diabetes. This risk was higher in both patients with diabetes with insulin-use (adjusted HR 1.51, 95% CI 1.31 to 1.74, p < 0.001) and without insulin-use (adjusted HR 1.22, 95% CI 1.13 to 1.33, p < 0.001), compared to patients without diabetes.
90-day readmission occurred in 19.4% (n = 20,542) of all patients. Patients with diabetes were more likely to experience 90-day readmission (HR 1.23, 95% CI 1.18 to 1.29, p < 0.001) compared to patients without diabetes. Patients with diabetes and insulin-use had an increased hazard rate for 90-day readmission (adjusted HR 1.46, 95% CI 1.34 to 1.59, p < 0.001), over and above patients with diabetes without insulin-use (adjusted HR 1.17, 95% CI 1.12 to 1.23, p < 0.001), compared to patients without diabetes (reference group).
Median length of stay was 8 days (IQR 5 to 15). Length of stay was increased by 10% in patients with diabetes (adjusted TR 1.10, 95% CI 1.08 to 1.11, p < 0.001).
Conclusion
Patients with diabetes undergoing colorectal resections are at a higher risk of 90-day mortality, prolonged length of stay and 90-day readmission, with insulin-use conferring additional risk. HbA1c is underutilised as a measure of risk stratification in the perioperative optimisation window for patients with diabetes requiring colorectal surgery. Awareness of these increased risks for patients with diabetes should be considered during perioperative optimisation of the patient, as well as in consent discussions and patient selection processes.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of ThesisFinal.pdf]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "ThesisFinal.pdf")