# Impact of diabetes mellitus and glycemic control on postoperative recurrence of perianal abscess: a retrospective, single-center study

## TL;DR

This study shows that diabetes and poor blood sugar control increase the risk of recurring perianal abscesses after surgery, and introduces a model to predict this risk.

## Contribution

A new risk prediction model for postoperative recurrence of perianal abscess incorporating diabetes and glycemic control is developed and validated.

## Key findings

- Diabetic patients had a 3.99 times higher risk of recurrence compared to non-diabetic patients.
- Each 1% increase in HbA1c was associated with a 1.50 times higher recurrence risk.
- The prediction model achieved strong discrimination with an AUC of 0.897.

## Abstract

To evaluate the independent effects of diabetes mellitus and the level of glycaemic control on postoperative recurrence of perianal abscess, and to develop and internally validate a clinically applicable risk prediction model.

We conducted a single-center retrospective cohort study of 232 consecutive adults who underwent primary incision and drainage for perianal abscess between June 2023 and March 2025. Detailed demographic, clinical and operative data were extracted from electronic records, including diabetes status and preoperative glycated hemoglobin (HbA1c). Candidate predictors (age, sex, BMI, lifestyle factors, abscess location, diabetes, HbA1c, inflammatory markers and intraoperative pus residue) were first entered into a LASSO regression for variable selection, followed by multivariable logistic regression. Model performance was assessed in terms of discrimination (area under the receiver operating characteristic curve, AUC), calibration (calibration plots and Hosmer–Lemeshow test) and clinical utility (decision curve analysis), with internal validation by bootstrap resampling (1,000 replicates).

Of the 232 patients (median age 47 years; 53.0% male), 68 (29.3%) experienced recurrence within 6 months. Compared with the non-recurrence group, patients with recurrence had a higher prevalence of diabetes (45.6% vs. 15.2%, p < 0.001) and higher median HbA1c levels (9.49% vs. 8.23%, p < 0.001), as well as higher age, BMI, a greater proportion of high intersphincteric abscesses and more frequent intraoperative pus residue. Multivariable analysis identified five independent predictors of recurrence: diabetes mellitus (OR 3.99, 95% CI 1.61–9.88, p = 0.003), HbA1c level (OR 1.50 per 1% increase, 95% CI 1.18–1.92, p = 0.001), BMI (OR 1.72 per kg/m2, 95% CI 1.43–2.06), age (OR 1.11 per year, 95% CI 1.06–1.16, p < 0.001), and low versus high intersphincteric abscess location (OR 0.36, 95% CI 0.14–0.97, p = 0.044). The resulting prediction model achieved an AUC of 0.897 (95% CI 0.814–0.957); after bootstrap internal validation, the optimism-corrected Harrell’s C-index was 0.892. Calibration plots showed good agreement between predicted and observed recurrence probabilities, and decision curve analysis indicated a consistent net clinical benefit across a wide range of threshold probabilities.

Diabetes mellitus and inadequate glycaemic control are important, independent risk factors for postoperative recurrence of perianal abscess, with a clear dose–response relationship between HbA1c and recurrence risk. The internally validated prediction model, which combines HbA1c with other readily available clinical variables, shows promise as a tool for early identification of high-risk patients and may support more personalized perioperative optimization and follow-up. External validation and impact studies are required before routine implementation in diverse clinical settings.