Predicting Infection Risk in Acute Compartment Syndrome: A Nomogram Model Based on Admission Blood Indicators

Introduction

Tibial fracture is one of the most common traumas of lower limbs caused by various injuries.¹ Acute compartment syndrome (ACS), as a severe complication after tibial fractures, usually occurs in 2–9% of patients who have suffered high-energy injuries.¹ The growing pressure within the compartment can rapidly reduce capillary blood flow, which may cause muscle necrosis or even amputation.^2,3 Currently, fasciotomy has been considered to be the most effective therapeutic method for ACS patients because it can quickly reduce intra-compartmental pressure and restore blood flow.^4,5 However, fasciotomy can lead to 20.8–36% of patients suffering from surgical site infection (SSI), which brings a huge economic and psychological burden to the families.^6–9

Ongoing articles have reported the predictors of SSI in ACS patients receiving fasciotomy. Yang found that open fractures, BMI, and smoking history were closely associated with SSI,⁹ but Yang did not establish the predictive model of SSI in ACS patients. Moldovan¹⁰ found the close relationship between serum biomarkers and Periprosthetic joint infection. To our knowledge, little research has focused on the predictors of SSI in ACS patients, let alone the crucial role of admission blood indicators in the diagnosis of SSI in ACS patients. Therefore, we aim to observe the role of admission blood indicators to predict SSI and then assess its potential biomarkers.

Patients and Methods

Ethics Statement

Our research was obtained by the institutional review boards of Affiliated Hospital of Hebei University of Engineering (2024–020-1) and exemption from informed consent as this is a retrospective study.

Patients

Data on ACS patients after fasciotomy in our hospital from Jan 2015 to Jan 2025. Based on whether ACS patients suffer from infection or not, they were separated into infection group (IG) and non-infection groups (NG). Diagnosis of infection is mainly based on the results of bacterial culture. The inclusion criteria were: 1) tibial fractures patients with ACS (ACS was defined as ∆P< 30 mmHg (∆P = diastolic arterial pressure-intra-compartmental pressure); 2) age ≥18 years; 3) ACS patients received fasciotomy. The exclusion criteria were: 1) patients with vascular injury; 2) patients without tibial fractures; 3) patients with insufficient data.

In our study, the patient characteristics, comorbidities, and admission laboratory results were obtained. The derived inflammatory indicators were included, such as NLR (neutrophils [NEU] count/lymphocytes count), MLR (monocytes count/lymphocytes count), PLR (platelets count/lymphocytes count), SII (platelets count × NEU count)/lymphocytes count), and SIRI (monocyte count × NEU count)/lymphocytes count)

Statistics

We used SPSS (version 27.0 SPSS Inc., Chicago, IL) to analyze the data with significance levels of p<0.05. When the continuous variables met normality criteria, we chose t-test, but if it did not met normality criteria, we used Mann–Whitney U-test. Regarding count data, we used the chi-square test. We used univariate and logistic regression analyses to analyze data. The cut-off values for continuous variables were used by the maximum Youden index (sensitivity+specificity-1) in the ROC curve analysis. Finally, we used R language software to establish the nomogram prediction model for SSI. Calibration curve and decision curve were used to assess the predicted probabilities and clinical application value, respectively.

Results

A total of 219 tibial fracture patients with ACS, including 38 with infection and 181 without infection, were included in this study. After selection based on the inclusion criteria and exclusion criteria, 14 patients with infection and 77 patients without infection were excluded (Figure 1). Finally, 24 patients with infection and 104 patients without infection were included in our study.

Figure 1 Flow diagram of included patients.

Table 1 showed that there is no difference in age, gender, BMI, mechanism of injury, open injury, blister, and muscle necrosis, as well as a history of smoking, alcohol, heart disease, hypertension, diabetes, and cerebral infarction in two groups (all p >0.05). However, we found that time from injury to admission (p<0.0001) and time from injury to surgery (p<0.0001) were closely related to SSI.

Table 1 Comparison of Characteristics of ACS Patients in Two Groups

Table 2 showed that there is no difference in the levels of hemoglobin (HGB), lymphocyte (LYM), platelet (PLT), red blood cell (RBC), PLR, and MLR in two groups (all p>0.05). However, we found that monocyte (MON, p<0.0001), NEU (p<0.0001), white blood cell (WBC, p<0.0001), total protein (TP, p<0.0001), C-reactive protein (CRP, p<0.0001), NLR (p<0.0001), SII (p<0.0001) and SIRI (p<0.0001) were closely related to SSI.

Table 2 Admission Laboratory Variables Associated with Infection in ACS Patients

Furthermore, we performed logistic regression analysis and found that the levels of NEU [p<0.0001, OR = 2.301, 95% CI (1.231, 4.353)], WBC [p<0.0001, OR = 1.669, 95% CI (1.069, 2.214)], CRP [p<0.0001, OR = 1.543, 95% CI (1.181, 2.103)] and time from injury to surgery [p<0.0001, OR = 1.213, 95% CI (1.077, 1.403)] were independent risk factors for ACS (Table 3).

Table 3 Logistic Regression Analysis of Laboratory Variables Associated with Infection in ACS Patients

We establish a nomogram prediction model according the findings of logistic regression analysis (Figure 2). ROC curve implies good discrimination ability [AUC = 0.995, 95% CI (0.990, 1.000), Figure 3] and the calibration curve shows well calibrated (p > 0.05, Figure 4). Decision curve analysis indicates our prediction model with good clinical benefits (Figure 5).

Figure 2 The nomogram prediction model for ACS.

Figure 3 The receiver operating characteristic curves of the nomogram.

Figure 4 The calibration curve of the nomogram.

Figure 5 The decision curve analysis of the nomogram.

Discussion

ACS, a common complication after tibial fractures, can result in tissue ischemia because of increased intra-compartment pressure, which may cause infection, amputation, or muscle necrosis.^1–3 Recently, limited articles reported the incidence and predictors of infection in ACS patients. A previous study has reported that open fractures, BMI, and smoking history were relevant to SSI,⁹ but it did not use the predictive model to observe the risk factors of SSI in ACS patients. Moldovan¹⁰ found the close relationship between serum biomarkers and Periprosthetic joint infection. As far as we know, few research has focused on the predictors of SSI in ACS patients, and ignored the important role of admission blood indicators in the diagnosis of SSI in ACS patients. Thus, the purpose of this article is to investigate whether admission blood indicators can predict SSI or not. In the current study, we found that various laboratory elements were related to SSI in ACS patients by univariate analysis. Furthermore, NEU, WBC, CRP and time from injury to surgery were closely related to SSI by logistic regression analysis. Our prediction model implies good discrimination ability and good clinical benefits.

NEU is the most common type of WBC in our body and plays a crucial role in the innate immune system.¹¹ LYM contributes to the adaptive immune system, including antibody production and cell-mediated immunity.¹² MON can release pro-inflammatory and pro-oxidative cytokines against pathogens.¹³ PLT is mainly responsible for regulating NEU and macrophage recruitment.¹⁴ Recently, inflammation ratios have been increasingly studied, including NLR, MLR, and PLR, These indicators have some advantages, such as whole blood count, low-cost, and routine clinical examinations.¹⁵ SII, stems from NEU, LYM, and PLT, can accurately reflect the immune and inflammatory levels of the body, which has been studied in the topic on predictors for inflammation diseases.¹⁶ SIRI, derived from NUE, LYM, and MON peripheral counts, can also represent inflammation.¹⁷ To our knowledge, the relationship between these indexes and SSI in ACS patients has not been studied.

In this study, our findings indicated that the levels of MON, NEU, WBC, TP, CRP, NLR, SII, and SIRI were associated with SSI in univariate analysis. Furthermore, logistic regression analysis indicated that the levels of NEU, WBC, and CRP were independent risk factors for SSI. Yang⁹ found, a significant correlation between NEU and SSI in univariate analysis, but in multivariate analysis, no significant correlation was found, which is inconsistent with our results and may be due to differences between samples. Therefore, we need large sample data for further research in future studies. As we know, CRP is an important indicator of inflammation levels in the body. Our study found that patients with elevated CRP have a 1.543 times higher risk of SSI after fasciotomy compared to patients with normal CRP levels.

Previous studies have demonstrated that smoking and high BMI were risk factors for SSI. Yang⁹ also found that these two factors are closely related to SSI, but our study did not find such results. We consider that this may still be related to the differences in the samples. We also investigated the correlation time from injury to admission, time from injury to surgery, and SSI. The results of univariate analysis showed that both were related to SSI, but the results of multivariate analysis only found a significant correlation between injury to surgery time and SSI. As is well known, 6–8 hours after injury is the early stage of treatment. When ACS occurs, high pressure in the compartment leads to tissue ischemia, and delayed treatment can cause muscle necrosis or even amputation. Therefore, the longer the time from injury to surgery, the longer the tissue ischemia time, and the higher the risk of natural tissue necrosis and infection.

Our prediction model to evaluate SSI in ACS patients. The ROC curve suggested good discrimination ability, and the calibration curve showed well calibrated. Decision curve analysis showed our nomogram prediction model with good clinical benefits. Although our study provides some innovative findings, a few limitations should be noted. First, single-center research and limited samples make it difficult to perform a subgroup analysis, which inevitably affects the reliability of the findings. Therefore, a multicenter, randomized controlled study with a larger sample size is required. Second, the inherent boundedness of data collection caused by a retrospective study causes some important inflammatory markers.

In conclusion, We found that the levels of NEU, WBC, CRP and time from injury to surgery were predictors for infection in ACS patients. Our nomogram prediction model can efficiently predict infection in ACS patients.