Diagnostic value of biochemical markers in prediction of perforation in acute appendicitis: a cross-sectional study

Abstract

Background:

Acute appendicitis, despite being the most common surgery, can result in increased morbidity and mortality in the case of perforated appendicitis. Identifying a perforation early can reduce the impact on the patient. Bilirubin, C-reactive protein (CRP), and white blood cell (WBC) count have been shown to indicate perforation in appendicitis.

Methods:

A cross-sectional analytical study was conducted to observe the association between preoperative bilirubin, CRP, and white cell count with the histological findings of either the presence or absence of perforation of the appendix. The eligibility criteria included all patients above 18 years with appendicitis who were managed surgically in the study hospital, Shree Birendra Hosptial, between 1 December 2021 to 30 May 2022. The study was conducted after the approval from the ethics committee.

Results:

Out of 150 patients, 112 had open surgery and 38 underwent laparoscopic appendectomy. One hundred thirty-four had non-perforated appendix and 16 cases had perforated appendix. The older age group was associated with appendiceal perforation more than the younger age group. Patients with a duration of symptoms less than 24 h and more than 72 h developed perforated appendicitis less frequently. The odds of encountering perforated appendicitis increased by 2.644 times per 0.972 rise in total bilirubin (P= 0.004) and the odds of encountering perforated appendicitis increased by 6.474 times per 1.868 rise in conjugated bilirubin (P= 0.003). There was no significant difference in total leucocyte count and neutrophil percentage among perforated and non-perforated groups.

Conclusion:

The diagnostic value of the CRP, total bilirubin, and conjugated bilirubin levels was strong and significant when used in combination rather than their individual performance in the diagnosis of perforated acute appendicitis.

Background

Highlights

• Prediction and early detection of perforated appendicitis, a surgical emergency, still remains a challenge.

• Various biochemical markers were measured to test their diagnostic value for perforated appendicitis.

• Diagnostic value of the CRP, total, and conjugated bilirubin were strong and significant when used in combination (AUC: 0.976).

• Older age, fever, jaundice, peritonitis, and presentation of symptoms of 24–72 h were also significantly associated with appendicular perforation.

Acute appendicitis (AA) is a common emergency affecting patients of all ages, sex, and geographical distribution with an incidence ranging from 100 to 223 cases per 100 000 individuals per year¹. Despite advances in diagnostic tools such as ultrasonography and multi-detector computed tomography (MDCT), diagnosis remains a challenge^2–4. Scoring systems developed to predict AA such as the Alvarado score and others based on symptom, sign, and laboratory findings, do not accurately assess the risk of appendiceal perforation, a surgical emergency^5–10. A systematic review showed that delays in undergoing surgery from the time of symptom onset resulted in statistically significant increases in rates of perforation (25.3– 40.6%)¹¹. Thus, there is an increasing need for markers to rapidly and accurately separate patients with an uncomplicated or complicated AA. White blood cell (WBC) count and C-reactive protein (CRP) are widely used biochemical markers in suspected appendicitis^12–15. When combined, they show high discriminatory power but alone they fall short¹⁶. Hyperbilirubinemia in combination with WBC count and CRP is shown to be a good predictive marker of perforated acute appendicitis^17–20.

This study aims to identify the predictive value of the above biochemical markers in the identification of perforated appendix for early and prompt recognition of perforated appendicitis.

Methods

Registration

This study has been registered on the Research Registry and the unique identifying number is researchregistry10313. This study is reported in line with the STROCSS guidelines²¹.

Ethical consideration

This study has followed the Declaration of Helsinki. Before the commencement of the study, ethical approval of the Nepal Health Research Council (Ref no: 1834) was taken. The participants of this study were made aware of the nature of the study and their consent was taken before including them in the study.

Study design and setting

This single tertiary care-centric prospective cross-sectional study spanned 6 months (from December 2021 to May 2022). Patients of age above eighteen years undergoing appendectomy (both open and laparoscopic) and giving consent for the study were included in this study. Patients aged less than eighteen, not giving consent, diagnosed with malignancies, in an immunocompromised state, or managed conservatively were excluded from the study (Fig. 1).

Figure 1.

Flow diagram of patient inclusion in the study.

Sample size and sampling technique

The total population sampling method was adopted for this study. In the span of six months, during the study period, we evaluated all the cases that were diagnosed with acute appendicitis. The diagnosis of acute appendicitis was done clinically and with correlation to USG (ultrasonogram) findings. However, CECT (Contrast-enhanced computed tomography) was not routinely performed and the diagnosis of perforated appendicitis was made intraoperatively, which was later confirmed by histopathological examination. We encountered a total of 189 patients during the study period and on comparing all the cases against the inclusion and exclusion criteria of this study, 150 patients were included as the participants of this cross-sectional study (Figure 1).

Study tool

A pre-formed structured questionnaire was used to collect data from the participants. The study tool was made in both the native language and English. This tool included slots of demographic details, past and present history of illness, diagnostic assessments, preoperative investigations, operative findings, and histopathology findings. The study tool is available as Supplementary File 1.

Variables

Independent variables of this study were age, sex, duration of symptoms, comorbidities, ALVARADO scoring, presence of peritonitis, jaundice, fever, operative procedure performed, and operative findings. Peritonitis is inflammation of the peritoneum, the serosal membrane lining the abdominal cavity and its visceral organs, typically caused by infectious (bacterial or fungal) agents but also possibly originating from noninfectious sources, presenting acutely either locally or throughout the abdomen²². The Merck Manual defines fever as an oral temperature exceeding 37.8°C (100.0°F), a rectal temperature surpassing 38.2°C (100.8°F), or a temperature higher than an individual’s typical daily baseline²³. Jaundice is characterized by a yellowing of body tissues due to an excess buildup of bilirubin²⁴. Dependent variables include operative findings and histopathology findings.

Analytical strategy

Data analysis was performed by using SPSS software version 20. The normality of the data was assessed by the Shapiro–Wilk test. Descriptive statistics were used to calculate the frequency and percentage of each categorical variable. For continuous variables, a median with an interquartile range (IQR) was used. Inferential statistics were used to see if any differences existed between the sub-groups of the variables. The χ² test along with Bonferroni correction was applied to know if any differences existed for categorical variables, while the Mann–Whitney U test along with binomial regression was applied for continuous variables. The receiver operating curve was used to determine the diagnostic performance of C-reactive protein, total bilirubin, and conjugated bilirubin for perforated appendicitis.

Results

A total of 150 cases diagnosed with acute appendicitis, with a male-to-female ratio of 1.41:1, were part of this study. The most common presenting complaint was nausea/vomiting, while jaundice and features suggesting peritonitis were the least common. More than half (52%, 78) of the patients who were diagnosed with acute appendicitis had a duration of symptoms of less than 24 hours. The total leucocyte count of the patients was found to be between 11 000 and 16 000 per microliter in more than two-thirds of the patients (69.3%, 104). Median (IQR) C-reactive protein, Total bilirubin, and Conjugated bilirubin were 4.70 (3.70–6.37), 0.90 (0.60–1.10), and 0.70 (0.30–0.90) respectively. Out of 150 patients included in this study, 112 (74.7%) underwent open appendectomy, and 38 (25.3%) underwent laparoscopic appendectomy. 16 (10.7%) had perforated appendicitis, and 134 (89.3%) had non-perforated appendicitis, which was initially based on intraoperative findings and later confirmed histopathologically. Cross-tabulation with chi-square test of categorical variables (age, sex, duration of symptoms, ALVARADO scoring, peritonitis, jaundice, and fever) with the dependent variable showed that age group, duration of symptoms, fever, jaundice, and peritonitis differed significantly. C-reactive protein, total bilirubin, and conjugated bilirubin tested against the dependent variable showed that total bilirubin and conjugated bilirubin had significant differences. Details of other patient demographics and clinical characteristics are presented in Table 1. The biochemical markers among the study participants are presented in Table 2 and the intraoperative and histopathological findings in Table 3.

Table 1.

Details of various patient demographics and clinical characteristics

	n (%)
Variables	N=150	Perforated	Non-perforated	P
Age group
18–30	81 (54.0)	5 (31.2)	76 (56.7)	0.028
31–45	31 (20.7)	2 (12.5)	29 (21.6)
46–60	18 (12.0)	4 (25.0)	14 (10.4)
> 60	20 (13.3)	5 (31.2)	15 (11.2)
Sex
Male	88 (58.7)	10 (62.5)	78 (58.2)	0.742
Female	62 (41.3)	6 (37.5)	56 (41.8)
Duration of symptoms
< 24 h	78 (52.0)	1 (6.3)	77 (57.5)	0.000
24–48 h	36 (24.0)	7 (43.8)	29 (21.6)
48–72 h	21 (14.0)	7 (43.8)	14 (10.4)
> 72 h	15 (10.0)	1 (6.3)	14 (10.4)
Nausea/vomiting
Yes	55 (36.7)	8 (50.0)	47 (35.1)	0.242
No	95 (63.3)	8 (50.0)	87 (64.9)
Fever
Yes	29 (19.3)	8 (50.0)	21 (15.7)	0.001
No	121 (80.7)	8 (50.0)	113 (84.3)
Jaundice
Yes	1 (0.7)	1 (6.3)	0 (0.0)	0.004
No	149 (99.3)	15 (93.7)	134 (100.0)
Peritonitis
Yes	1 (0.7)	1 (6.3)	0	0.004
No	149 (99.3)	15 (93.7)	134 (100.0)
ALVARADO score
Up to 4	6 (4.0)	0 (0.0)	6 (4.5)	0.673
5 and 6	33 (22.0)	4 (25.0)	29 (21.6)
7 or more	111 (74.0)	12 (75.0)	99 (73.9)

Table 2.

Biochemical markers in study participants

	n (%)
Variables	N=150	Perforated	Non-perforated	P
Leucocyte count (per microliter)
<11 000	32 (21.3)	2 (12.5)	30 (22.4)	0.548
11,000–16,000	104 (69.3)	13 (81.2)	91 (67.9)
> 16 000	14 (9.3)	1 (6.3)	13 (9.7)
Neutrophils
< 75%	20 (13.3)	2 (12.5)	18 (13.4)	0.934
76–85%	107 (71.3)	12 (75.0)	95 (70.9)
> 85%	23 (15.3)	2 (12.5)	21 (15.7)
C-reactive protein (mg/l)
< 3	25 (16.7)	2 (12.5)	23 (17.2)	0.479
≥ 3	125 (83.3)	14 (87.5)	111 (82.8)
Total bilirubin (mg/dl)
< 1.2	121 (80.7)	9 (56.3)	112 (83.6)	0.016
≥ 1.2	29 (19.3)	7 (43.8)	22 (16.4)

Table 3.

Intra-operative and histopathology findings

	N=150
Variables	n (%)
Surgery performed
Open appendectomy	112 (74.7)
Laparoscopic appendectomy	38 (25.3)
Intraoperative findings
Non-perforated appendicitis	134 (89.3)
Perforated appendicitis	16 (10.7)
Histopathology findings
Non-perforated appendicitis	134 (89.3)
Perforated appendicitis	16 (10.7)

Statistically significant categorical values were then treated with Bonferroni correction which showed that the older age group (above 46 years) more frequently developed perforated appendicitis in comparison to the younger age group (18–30 years). Patients with a duration of symptoms lesser than 24 h developed perforated appendicitis less frequently compared to the duration of symptoms of 24–72 h. Statistically significant continuous variables were treated with binomial logistic regression, which showed that the odds of encountering perforated appendicitis increased by 2.644 times per 0.972 rise in total bilirubin and odds of encountering perforated appendicitis increased by 6.474 times per 1.868 rise in conjugated bilirubin. Details are shown in Tables 4 and 5.

Table 4.

Bonferroni correction for Intraoperative findings × Age group and Duration of symptoms

Variables	Z score	df	Corrected P	P
Perforated
Age group
18–30	−3.6	1	0.00625	0.0003
31–45	−1.3	1		0.1936
46–60	2.1	1		0.0357
> 60	2.9	1		0.0037
Non-perforated
Age group
18–30	3.6	1	0.00625	0.0003
31–45	1.3	1		0.1936
46–60	−2.1	1		0.0357
> 60	−2.9	1		0.0037
Perforated
Duration of symptoms
< 24 h	−7.3	1	0.00625	0.0000
24–48 h	3.2	1		0.00137
48–72 h	4.8	1		0.0000
> 72 h	−0.6	1		0.54851
Non-perforated
Duration of symptoms
< 24 h	7.3	1	0.00625	0.0000
24–48 h	−3.2	1		0.00137
48–72 h	−4.8	1		0.0000
> 72 h	0.6	1		0.54851

df, degrees of freedom.

Table 5.

Binomial logistic regression, univariate analysis

				95% CI
Variables	Coeff	P value	OR	Lower	Upper
Total bilirubin	0.972	0.004	2.644	1.366	5.117
Conjugated bilirubin	1.868	0.003	6.474	1.854	22.607

Coeff, coefficient beta; OR, odds ratio.

Receiver operating curve analysis was performed to determine the diagnostic performance of the C-reactive protein, Total bilirubin, and Conjugated bilirubin. The obtained area under the curve was statistically significant for total bilirubin and conjugated bilirubin. The cut-off value for total bilirubin is 0.7450, which has a sensitivity of 87.5% and specificity of 40.3%. While the cut-off value for conjugated bilirubin is 0.5500, which has a sensitivity of 87.5% and specificity of 34.3%.

C-reactive protein, total bilirubin, and conjugated bilirubin were treated with logistic regression, and the resultant combined predicted probabilities were used to construct a receiver operating curve, which showed improved diagnostic value (area: 0.976, 95% CI: 0.954–0.999, P value: <0.001). The cut-off value for the combined variables at 0.1141 has a sensitivity of 100.0% and specificity of 89.6%. Details are shown in Table 6, Figures 2 and 3.

Table 6.

Receiver operating curve analysis

		95% CI
Variables	Area under curve	Lower	Upper	P
C-reactive protein	0.601	0.449	0.754	0.185
Total bilirubin	0.712	0.566	0.858	0.006
Conjugated bilirubin	0.683	0.528	0.838	0.079
Combined Predicted Probabilities (CRP, total and conjugated Bilirubin)	0.976	0.954	0.999	0.000

CRP, C-reactive protein.

Figure 2.

Receiver operating characteristic (ROC) curve analysis of the diagnostic performance of C-reactive protein, total bilirubin, and conjugated bilirubin.

Figure 3.

Receiver operating characteristic (ROC) curve analysis of the diagnostic performance of the combined predicted probabilities of C-reactive protein, total bilirubin, and conjugated bilirubin.

Discussion

Early diagnosis of acute appendicitis is not always as easy as it appears to be. Despite the advent of advanced testing using CT and USG, appendicitis poses a serious dilemma to surgeons regarding the decision to observe the patient until the diagnosis is confirmed or to operate early to prevent complications like perforation and peritonitis. Furthermore, there are serious clinical and financial costs incurred by the patients undergoing appendectomy. Some studies have shown a high rate of negative appendectomies^25–27. It is, therefore, important to differentiate early appendicitis from other causes of acute abdominal pathologies.

In the present study, we aimed to determine the value of biochemical markers CRP, bilirubin, and WBC as diagnostic markers of perforated acute appendicitis. The common clinical presentation among patients diagnosed with acute appendicitis was nausea and vomiting followed by fever. Features of jaundice and peritonitis were present in only one of the patients. This study also showed an increased rate of appendiceal perforation among patients presenting with fever compared to those without fever²⁸. Similarly, those with an ALVARADO score of 7 or more had a higher incidence of perforation (12, 8%). This highlights the importance of clinical scoring systems like ALVARADO as an aid to biochemical markers in risk stratification of patients with suspected appendicitis²⁹.

Among the total of 150 patients who were included in the study 112 (74.67%) had an emergency open surgery and 38 (25.33%) had a laparoscopic appendectomy. Those patients who were managed conservatively and did not give consent were excluded from the study. Among these patients, the postoperative histopathological evaluation showed that 134 (89.33%) had acute appendicitis while only 16 (10.66%) cases had perforated appendix. These findings were similar to the study conducted by D’Souza and colleagues, they had simple appendicitis in 89 (63%) cases, and perforated appendicitis in 19 (14%) cases. Both of these studies showed that there is a higher incidence of simple appendicitis as compared to perforated appendicitis.

Our study showed that out of a total of 150 patients, 88 (58.67%) were males and 62 (41.33%) were females. Most of them (81, 54%) belonged to the 18–30 years age group. These findings were similar to a few other studies conducted in Nepal, which showed the highest incidence amongst the 21–30 years age group with a predominance of male (60.42%) over females (39.58%)³⁰. This study also corroborates the higher incidence of perforation with increasing age (above 46) as noted in previous studies³¹.

Our study showed that appendiceal perforation was highest in patients who presented between 24 and 72 h of symptom onset compared to those who visited the hospital earlier or later and was found to be statistically significant. This shows the correlation of perforation of appendicitis with increased duration of symptoms. The time between symptom onset and treatment, in the form of pre-hospital delay or the hospital setting delay, is a potentially modifiable factor associated with the risk of ruptured appendicitis, as seen in other studies^32,33.

In our study, all patients underwent laboratory testing including WBC count, bilirubin (total and conjugated) and CRP as a part of the institutional preoperative protocol. The median CRP level in our study was higher in a perforated group compared to the non-perforated group [5.50 (3.73–28.93) vs. 4.70 (3.70–6.15)] but this finding was not statistically significant (P=0.185). These findings were similar to a study conducted in Nepal, which showed Negative CRP was found in 18.52% of patients among whom 14.81% of patients were histopathologically proved as acute appendicitis³⁴. The study concluded that the level of CRP with preoperative findings and histopathology reports was not statistically significant, but the majority of patients with significantly raised CRP levels (>85 mg/dl) had severely inflamed or gangrenous appendices³⁴. Other studies have shown that levels of bilirubin and CRP were significantly increased in the early diagnosis of acute appendicitis^35,36. However, total serum bilirubin was significantly higher in cases with perforated appendicitis in comparison to those without perforation. Likewise, the level of conjugated bilirubin was also significantly higher in the former group. Similar findings were also reported in studies conducted by Valon and colleagues, Motie and colleagues, and Alfehaid and colleagues.^37–39. In our study, there was no significant difference in total leucocyte count and neutrophil percentage among perforated and non-perforated groups. However, other studies have shown that white cell count has moderate diagnostic accuracy with the neutrophil-to-lymphocyte ratio being a better predictor of both diagnosis and severity of acute appendicitis^14,40. Inflammatory response variables like WBC count, percentage of neutrophils, and CRP had a higher propensity of discriminating perforated from non-perforated appendicitis although statistically insignificant. A study conducted by McGowan et al. ⁴¹ showed that CRP has a higher sensitivity than other blood test results, mainly because of the relatively low normal value cut-off. Results of this study do not confidently show that blood tests are diagnostic tools⁴¹. However, if considered together with the clinical picture, they can be useful in prioritizing patients and whether a patient is suitable for conservative management or not. The diagnostic value of these markers in acute appendicitis has been extensively studied and each has been shown to fall short when used as a single diagnostic marker alone. However, when combined they show high discriminatory power^12,14,36,41.

Our study is not free from limitations. It was conducted in a single centre, and hence the results may not be generalized to the entire population. There was no standardized protocol for the timing of blood sampling for the biochemical markers or the timing of surgery from presentation. The correlation between biochemical markers with intraoperative findings like the site of perforation, amount of peritoneal collection and extent of peritonitis were not included in our study.

Conclusion

Diagnostic value of the CRP, total bilirubin, and conjugated bilirubin levels were strong and significant when used in combination rather than their individual performance in the diagnosis of perforated acute appendicitis. Older age, fever, jaundice and peritonitis at presentation, and duration of symptoms of 24–72 h were the other risk factors that were significantly associated with appendicular perforation. Thus, biochemical markers when used in conjunction with clinical parameters seem to yield the highest efficacy in distinguishing perforated appendicitis from non-perforated.

Ethical approval

Ethical clearance for this study was received from Nepal Health Research Council (NHRC) Ref no. 1834.

Consent

Consent to participate in the study was taken before the data collection.

Source of funding

This article didn’t receive any grants.

Author contribution

S.B., S.G., and A.M.B. were involved in conceptualization of the study. O.S., N.T., A.T., and B.D.P. were involved in data curation. S.B., O.S., N.T. and K.M. did the formal analysis. S.B., O.S., N.T., A.T., B.D.P. were involved in performing literature review and initial manuscript drafting. S.G., A.M.B. and M.D. revised the manuscript. All the authors have read and approved the final version of the manuscript.

Conflicts of interest disclosure

The authors declare no conflicts of interest.

Research registration unique identifying number (UIN)

1. Name of the registry: Research Registry (http://www.researchregistry.com).

2. Unique Identifying number or registration ID: researchregistry10313.

3. Hyperlink to your specific registration (must be publicly accessible and will be checked). https://www.researchregistry.com/registernow#home/registrationdetails/66477c1137c444002710ee61/.

Guarantor

Sunil Basukala.

Data availability statement

Curated data that were analyzed are available from the corresponding author on reasonable request.

Provenance and peer review

Not commissioned, externally peer-reviewed.

Transparency statement

We affirm that the submitted manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.