Results

Demographics and comorbidities

A total of 196 patients were identified for the final analysis, and the baseline characteristics were summarised in Table 1.

Characteristics | Frequency (n) | Percentage (%)
Gender (n=196)
Male | 148 | 75.5
Female | 48 | 24.5
Composite comorbiditiesa(n=167)
Diabetes mellitus | 105 | 53.5
Hypertension | 129 | 65.8
Vascular procedures (n=196)
Open procedures (n=119) | 119 | 60.7b
Open amputation | 36 | 30.3
Open endarterectomy | 21 | 17.6
Bypass | 25 | 21.0
Others | 37 | 31.1
Minimally invasive procedures (n=77) | 77 | 39.2b
Angioplasty ± stenting | 53 | 68.8
Endovascular aneurysm repair (EVAR) | 14 | 18.2
Others | 10 | 13.0

The cohort was characterised by age profile, with a mean age of 69 years (±15.38). The population was predominantly male, comprising 148 (75.5%) males and 48 (24.5%) females. A high prevalence of metabolic comorbidities was observed, with 167 (85.2%) of patients presenting with composite comorbidities, defined as the presence of diabetes mellitus, hypertension, or both. Individually, hypertension affected 129 (65.8%) of the cohort, while diabetes mellitus was present in 105 (53.5%). The sum of the individual comorbidity frequencies exceeds the composite prevalence due to substantial overlap between the two conditions, with many patients presenting with both hypertension and diabetes mellitus simultaneously.

Surgical procedures

As detailed in Table 1, surgical interventions were categorised according to operative approach. Open procedures accounted for 119 (60.7%) of all operations. Within this subgroup, open amputations were the most common intervention (36, 30.3%), followed by bypass surgery (25, 21.0%) and endarterectomy (21, 17.6%). Other open procedures, including debridement, incision and drainage, and open repair, comprised 37 (31.1%) of the open cases.

Minimally invasive procedures were performed in 77 (39.2%) of the cohort. The majority consisted of angioplasty with or without stenting (53, 68.8%), while endovascular aneurysm repair (EVAR) accounted for 14 (18.2%). An additional 10 (13.0%) of the cases involved miscellaneous endovascular or minimally invasive procedures, including bone biopsy, sclerotherapy, mechanical thrombectomy, and embolisation.

Anaemia prevalence and severity across the perioperative window

The distribution of anaemia severity at preoperative and postoperative time points classified according to WHO criteria is summarised in Table 2 and Figure 2.

Clinical Stages | No Anaemia, n (%) | Mild Anaemia, n (%) | Moderate Anaemia, n (%) | Severe Anaemia, n (%) | Total Anaemia, n (%)
Preoperative | 48 (24.5) | 79 (40.3) | 60 (30.6) | 9 (4.6) | 148 (75.5)
Postoperative | 40 (20.4) | 38 (19.4) | 99 (50.5) | 19 (9.7) | 156 (79.6)

Figure 2: Distribution of Anaemia Prevalence and Severity in the Preoperative and Postoperative Periods (N=196).

Preoperatively, 48 (24.5%) of the cohort presented without anaemia. Within the anaemic population, 79 (40.3%) were classified as mild and 60 (30.6%) were moderate, while only nine (4.6%) presented with severe anaemia.

Following surgical intervention, a significant downward shift in Hb concentrations was observed, with the cohort mean decreasing from a preoperative baseline of 115 ± 18.9 g/L to a postoperative mean Hb of 105 ± 17.9 g/L. This physiological shift was characterised by a distinct migration toward more advanced anaemia categories. While the proportion of patients without anaemia remained relatively stable (40, 20.4%), the prevalence of mild anaemia halved to 38 (19.4%). Conversely, the moderate anaemia cohort increased substantially to 99 (50.5%), and the prevalence of severe anaemia exhibited a two-fold increase, reaching 19 (9.7%) in the postoperative period.

Primary clinical outcomes

The primary composite adverse outcome occurred in 158 (80.6%) patients. Within this group, 140 (71.4%) patients experienced a prolonged hospitalisation (seven or more days), 50 (25.5%) required an unplanned 30-day readmission, and 16 (8.2%) suffered 30-day mortality. As some patients experienced more than one of these complications, the total number of discrete events recorded was 206. A detailed breakdown of these outcomes is provided in Table 3.

Outcome Measure | Frequency (n) | Percentage (%)a
Prolonged Hospitalisation (≥7 Days) | 140 | 71.4
Unplanned 30-Day Readmission | 50 | 25.5
30-Day Mortality | 16 | 8.2
Composite Adverse Outcome | 158 | 80.6

Subgroup analysis and clinical outcomes

Exploratory subgroup comparisons were conducted to evaluate the association of baseline characteristics and the 30-day composite adverse outcome utilising Pearson's chi-square test. The results were summarised in Table 4.

 | Chi-Square Statistic (X2) | Degrees of Freedom (df) | p-value | Effect Size (Cramer’s V)
Gender | 0.30 | 1 | 0.583 | 0.039
Composite Comorbidities | 0.49 | 1 | 0.483 | 0.050
Vascular Procedures | 11.26 | 1 | <0.001 | 0.240

Comparative model performance

The comparative performance of the three evaluated models is summarised in Table 5.

Models | p-value | Akaike Information Criterion (AIC) | Nagelkerke R2 | Area Under the Receiver Operating Characteristic Curve (AUC)
Preoperative Severity Model | 0.005 | 188.10 | 0.10 | 0.67
Postoperative Severity Model | 0.001 | 184.10 | 0.13 | 0.69
Persistent Perioperative Model | 0.025 | 191.74 | 0.04 | 0.60

The postoperative anaemia model demonstrated the highest predictive accuracy and model fit (AIC=184.10; R2=0.13; AUC=0.69). In comparison, lower predictive utility was observed for both the preoperative anaemia model (AIC=188.10; R2=0.10; AUC=0.67) and the persistent perioperative anaemia model (AIC=191.74; R2=0.04; AUC=0.60).

Association between postoperative anaemia and 30-day adverse outcomes

Analysis of the postoperative anaemia model demonstrated a significant association between the severity of anaemia and the 30-day composite adverse outcome as detailed in Table 6.

Postoperative Anaemia Severity | p-value | Odds Ratios (ORs) | 95% Confidence Intervals (CIs)
No Anaemia (Reference) | - | 1.00 | -
Mild Anaemia | 0.749 | 1.18 | 0.45-3.00
Moderate Anaemia | 0.002 | 4.31 | 1.75-10.63
Severe Anaemia | 0.035 | 9.69 | 1.17-80.41

While mild anaemia showed no significant impact on clinical outcomes (p=0.749; OR: 1.18; 95% CI: 0.45-3.00), a strong correlation was observed at higher severity levels. Specifically, moderate anaemia was associated with a four-fold increase in risk (p=0.002; OR: 4.31; 95% CI:1.75-10.63), while severe anaemia resulted in a nearly 10-fold escalation in the odds of an adverse event (p=0.035; OR: 9.69; 95% CI: 1.17-80.41).