In this study, we developed and validated a novel scoring model for GPA in adults using a combination of patient clinical characteristics, laboratory tests, and abdominal CT results. Univariate and multivariate logistic regression analyses identified seven independent predictors of GPA: WBC, lymphocyte, D-dimer, glucose, and albumin levels from laboratory tests, as well as the maximum outer diameter of the appendix and the presence of appendix fecalith as detected through abdominal CT scans. A nomogram was constructed and validated using these predictors to assess the risk of GPA in adults. The laboratory-CT model exhibited strong diagnostic accuracy in predicting the occurrence of gangrenous appendicitis in adults.
Acute appendicitis is a common abdominal emergency in adults worldwide, often requiring appendectomy as a surgical intervention [1]. Traditionally, surgeons have relied on clinical symptoms, laboratory findings (such as elevated white blood cell count), and abdominal CT scans to guide the decision-making process for surgical intervention in affected patients. Recent research has demonstrated the efficacy of antibiotic therapy for uncomplicated appendicitis, while complicated cases typically still require appendectomy [3–9]. Accurately identifying complicated appendicitis is crucial for guiding appropriate treatment decisions and avoiding unnecessary surgeries. However, there is limited literature on predictive models using laboratory tests and abdominal CT scans to anticipate complicated appendicitis in adult patients.
Previous studies have reported various indicators associated with GPA [20–32]. WBC is commonly used as a blood marker for diagnosing appendicitis [21, 22], while lymphocyte levels are inversely correlated with the severity of appendicitis inflammation, typically reduced in cases of gangrenous appendicitis [23, 24]. Our study confirmed WBC and lymphocytes as independent predictors of GPA, consistent with existing literature.
Research has indicated a bidirectional relationship between the coagulation system and inflammatory response [25]. D-dimer, a degradation product of fibrin, has been studied in the context of non-traumatic acute abdomen, suggesting its potential to replace white blood cells in diagnosing this condition [26]. Furthermore, Blunt et al. found that D-dimer demonstrates high specificity but low sensitivity in diagnosing appendicitis [26]. Previous studies have also suggested a potential link between D-dimer levels and the severity of appendicitis [27]. Elevated blood sugar levels have been noted as a response to inflammatory triggers and physical stress, creating a favorable environment for bacterial growth and worsening inflammation in the appendix [28]. Additionally, albumin (ALB), a negative acute-phase reactant produced by the liver, decreases in response to inflammation, with implications for inflammation severity, disease prognosis, and mortality [29]. In our research, these markers were identified as independent predictors of GPA through multivariate analysis, supporting the pathophysiological mechanisms of the disease and confirming previous research findings.
Previous studies have suggested that the maximum diameter of the appendix and the presence of fecalith are linked to the failure of non-surgical treatments for uncomplicated appendicitis [30]. Outlet obstruction of the appendix is widely acknowledged as a primary cause of appendicitis [31], with fecalith leading to persistent blockage of the appendix lumen and subsequent increase in intracavitary pressure, promoting the progression of inflammatory processes. Additionally, it is theorized that the diameter of the appendix may enlarge in response to elevated intracavitary pressure, potentially leading to complications such as gangrenous appendicitis and perforation [32]. Therefore, our research suggests that the presence of fecalith and an enlarged appendix diameter are strongly associated with the development of gangrenous appendicitis and the failure of non-surgical interventions.
In our study, we analyzed various clinical characteristics of patients, including age, gender, duration of abdominal pain, history of hypertension and diabetes, history of abdominal surgery, and history of preoperative antibiotic treatment. However, statistical analysis revealed that these factors were not independent predictors of GPA. Instead, our scoring model incorporated seven variables - WBC, lymphocytes, D-dimer, blood glucose, albumin, maximum diameter of the appendix, and presence of fecalith - which provided objective and reproducible data for easy utilization.
The evaluation of our scoring model demonstrated its strong discriminative ability in accurately predicting GPA. External validation using patient data from the Fifth Affiliated Hospital of Sun Yat-sen University showed no significant variations in baseline characteristics between the training and test sets, confirming their suitability as training and validation sets. The nomogram consistently performed well in the verification set, indicating successful validation across diverse populations. This well-calibrated and distinctive nomogram, validated in both training and testing sets, serves as a valuable tool for identifying adult GPA and guiding surgical decision-making in appendicitis patients.
Moreover, the model's objectivity and reproducibility make it applicable in scenarios where patients may have difficulty providing accurate medical histories or undergoing abdominal examinations due to conditions like impaired consciousness or communication limitations.
Despite its strengths, our model acknowledges limitations in result interpretation. The retrospective design of the case-control study introduces the potential for selection bias, and the use of data from only two centers may restrict the generalizability of our findings to a broader population at risk of GPA. Further research with larger and more diverse samples is warranted to address these limitations.