Diagnostic value of laboratory results in children with acute appendicitis

Introduction

Acute appendicitis (AA) is the most prevalent cause of abdominal pain requiring emergency operation among children [1]. Diagnosis can be delayed due to atypical patient complaints and physical examination findings [2]. Late diagnosis increases the risk of perforation due to which morbidity, mortality, and cost of treatment increase [3]. Despite advanced laboratory and radiologic examinations, the correct diagnosis rate is between 72 and 94% [4].

Alvarado score is a scoring system which is reported to have 78–82% accuracy in the diagnosis of AA [5]. In this scoring system, evaluation is made according to the clinical findings of patients, physical examination and laboratory findings [6]. Leukocyte increase (>10,000/mm³) and leukocyte left shift (>75%) are laboratory findings used in the Alvarado score. Complete blood count (CBC) is a cheap and easily accessible test, which can be performed in almost all centres. Additionally, CBC helps evaluate white blood cells (WBC), neutrophile and lymphocyte counts, red cell distribution (RDW) and mean platelet volume (MPV).

In this study, it was aimed to determine the contribution of Alvarado scoring system and neutrophil-to-lymphocyte ratio (NLR), RDW, MPV, C-reactive protein (CRP) and procalcitonin (PCT) values to the diagnosis of the patients diagnosed with AA histopathologically. Moreover, the usability of these values in the separation of perforated and non-perforated cases was emphasized.

Materials and methods

This study was conducted in compliance with the fundamentals of the Declaration of Helsinki and approval was acquired from the Fırat University Non-Interventional Research Ethics Board (Decision number: 06, Date: 22/11/2018). Records of 201 patients between the ages of 0–16 who were operated for AA between January 2017 and November 2018 in the pediatric surgery clinic of a university hospital were retrospectively analysed. The record form was created based on patient age, sex and the WBC, neutrophil and lymphocyte count, NLR, RDW, MPV, CRP and PCT values of the patients checked at the time of first admittance. Alvarado score was used as the clinical scoring system in the definition of AA (Table 1). Patients were divided into three groups according to the Alvarado score: Group 1; ≤4 points (low probability appendicitis), Group 2; 5–7 points (medium risk) and Group 3; 8–10 points (high risk). Moreover, the patients were histopathologically divided into two groups as AA and without appendicitis. Patients diagnosed with AA were subdivided into two groups as perforated (complicated) and non-perforated (noncomplicated). The patients who used antibiotics or were administered chemotherapy or steroids within the last 15 days due to any reason and those with diagnosed haematologic diseases were excluded from the study.

Results

Out of the 201 patients included in the study, 123 (61.2%) were male and 78 (38.8%) were female. The median age value was 12 (IQR:8–14) years. Histopathologic distribution of the patients according to the Alvarado score is presented in Table 2.

WBC, neutrophil, NLR and CRP values of patients histopathologically diagnosed with appendicitis were higher than those of patients without appendicitis (Table 3). Area under the ROC curve (AUC) and cutoff values for NLR, CRP, PCT and Alvarado score of the patients according to the ROC curve are presented in Table 4 and Figure 1.

Figure 1:

ROC curves for NLR, CRP, PCT and Alvarado score of patients histopathologically with and without appendicitis.

NLR, RDW, CRP and PCT values of the complicated patients were higher compared with those of the noncomplicated ones (p<0.05, Table 3). WBC and neutrophil values of complicated patients were higher than those of noncomplicated patients. However, the difference was not statistically significant (p>0.05, Table 3). AUC and cutoff values for NLR, RDW, CRP, PCT and Alvarado score of patient groups according to the ROC curve are presented in Table 5 and Figure 2.

Figure 2:

ROC curves for NLR, RDW, CRP, PCT and Alvarado score of noncomplicated and complicated patients.

Discussion

Although imaging methods are more easily accessible, there can be a delay in diagnosis at a rate of 30% [2]. It has been reported that a normal appendix is encountered at a rate reaching 15.7% among patients operated for AA [7]. In the present study, the male/female ratio was 1.57. The negative appendectomy rate was 7.5% (n=15). The existence of atypical symptoms and findings, especially among children, can cause delays in diagnosis, ultimately increasing morbidity and mortality [2]. Studies have reported on parameters such as NLR, MPV and RDW to increase the diagnostic accuracy rate in AA [].

It has been stated that NLR can be a better marker than neutrophil and CRP [8]. Different cutoff values have been emphasized in studies (1.77, 4, 5, 5.55) [7], [11], [12]. In literature, sensitivity and specificity were reported as 70.9% and 70–75% for NLR, respectively [11], [12]. In our study, sensitivity for the most appropriate NLR cutoff value according to the ROC curve was found compatible with that in the literature but specificity was found lower. When checked for noncomplicated and complicated cases, sensitivity and specificity were 73 and 52% in the literature [12]. In our study, sensitivity was higher and specificity was lower for the most appropriate NLR cutoff value according to the ROC curve in noncomplicated and complicated patients. NLR was statistically significant (p=0.025) for patients diagnosed with AA according to the ROC curve. NLR was not statistically significant for complicated cases according to the ROC curve (p=0.094). Thus, it was considered that although NLR contributes to the diagnosis of AA, it does not contribute to the distinction of complicated and noncomplicated cases.

In the conducted studies, it has been shown that RDW level changes in inflammatory intestinal diseases, acute pancreatitis, bacteraemia, sepsis and septic shock. Moreover, it was shown that there is a strong relationship between inflammatory markers such as RDW and CRP [13]. In a study conducted by Narci et al., it was reported that although RDW level was within the normal limits in a group with AA, it was significantly lower [10]. In a study conducted by Ertekin et al., it was reported that RDW level was higher among cases with AA than in the control group (p=0.012), but the difference was not statistically significant between the noncomplicated and complicated cases (p=0.94) [14]. In contrast, it has been reported that RDW does not have a significant value for AA diagnosis but significantly higher value for the complicated appendicitis group [9], [15]. In our study, although the median value of RDW was higher among the cases diagnosed with AA, it was not statistically significant (p=0.7); however, a high level of RDW for complicated cases was statistically significant (p<0.001). Increased RDW values are reported to be related to nutritional deficiencies [16]. Thus, the significant increase in the RDW value among the complicated cases may be related to the nutritional deficiency that is related to the loss of appetite continued for a longer duration compared with that for the noncomplicated cases. In our study, it was considered that the RDW value did not contribute to the early diagnosis of AA but was a supportive parameter in the determination of complicated cases with delayed diagnosis.

One of the thrombocyte indexes, MPV is a bioindicator related to thrombocyte activation and morphology [17]. MPV is the measurement of the average thrombocyte size in the peripheral blood. In this study, it was shown that there is a relationship between MPV and some inflammatory cases such as inflammatory intestinal diseases [17]. While it was reported that MPV decreases in AA [9], it was reported to increase in a different study [15]. In a study conducted by Uyanik et al., it was stated that MPV does not have a diagnostic value in pediatric AA cases [18]. In our study, MPV was not statistically significant in AA diagnosis (p=0.536) and the distinction of complicated cases (p=0.795).

CRP, the acute phase reactant, begins to increase between 8 and 12 h of inflammation and reaches its peak between 24 and 48 h [19]. In the conducted studies, a significant relationship was found between elevated CRP and AA diagnosis. Moreover, it was reported that CRP specificity is higher in complicated cases [20]. In literature, sensitivity and specificity were 50 and 80% for cases diagnosed with AA, respectively [21]. In our study, sensitivity and specificity were found higher compared with those in the literature (Table 4). Moreover, in literature, sensitivity and specificity were 85 and 73% for complicated cases, respectively [22]. While sensitivity was compatible with that in the literature, specificity was found to be higher in our study. CRP was significant in cases histopathologically diagnosed with AA and complicated cases (p<0.001). Thus, CRP is an effective bioindicator in the diagnosis of AA and in the distinction of noncomplicated–complicated cases.

Procalcitonin is an inflammatory bioindicator for which serum level does not change in cases of viral infections, autoimmune diseases and surgical trauma [23]. In a study conducted by Yu et al., it was reported that the highest accuracy rates in the diagnosis of AA lie with CRP, WBC and PCT [24]. PCT serum level may be low in the early stage of infection and in local infections [25]. In literature, sensitivity and specificity of PCT were reported to be 73.4 and 94.6% for the diagnosis of AA [26]. In our study, PCT level was not statistically significant for the diagnosis of AA (p=0.144). This may be due to the limitation of infection within the appendix tissue in noncomplicated cases. There are studies reporting that PCT has high sensitivity and specificity (62%, 94%, respectively) in the diagnosis of complicated appendicitis [24], [27]. In compliance with the literature, PCT level in this study was statistically significant in defining complicated cases (p<0.001). Thus, it was considered that PCT does not contribute to the diagnosis of AA but contributes to the distinction of complicated cases.

The limitations of the present study include its retrospective design and being a single-site study. Prospective and multi-site studies would enable acquisition of more accurate information.

In conclusion, NLR and CRP are bioindicators that can be used in addition to WBC and neutrophil dominance of laboratory parameters of the Alvarado score in the diagnosis of AA. CRP, PCT and RDW—despite its low sensitivity–specificity—are effective haematologic parameters in the distinction of complicated cases. MPV is not an effective parameter in the diagnosis of AA and in the distinction of complicated cases.