Is acute appendicitis still misdiagnosed?

OBJECTIVE: The optimal diagnostics and treatment of acute appendicitis continues to be a challenge. A false positive diagnosis of appendicitis may lead to an unnecessary operation, which has been appropriately termed negative appendectomy. The aim of our study was to identify the effectiveness of preoperative investigations in preventing negative appendectomy.
METHODS: A retrospective study was performed on adult patients who underwent operation for suspected acute appendicitis from 2008 to 2013 at Vilnius University Hospital Santariskiu Klinikos. Patients were divided into two groups: group A underwent an operation, where appendix was found to be normal (non-inflamed); group B underwent an appendectomy for inflamed appendix. Groups were compared for preoperative data, investigations, treatment results and pathology findings.
RESULTS: 554 patients were included in the study. Preoperative laboratory tests results of hemoglobin, hematocrit concentrations and white blood cell count were significantly higher in group B (p<0.001). Ultrasonography was performed for 78 % of patients in group A and 74 % in group B and did not provide any statistically significant results. Comparing Alvarado score results, there were more patients with Alvarado score less than 7 in group A than in group B. In our large series we could find only four independent risk factors, and they could only account for 24 % of cases.
CONCLUSIONS: In summary, acute appendicitis is still often misdiagnosed and the ratio of negative appendectomies remains rather high. Additional investigations such as observation and computed tomography should be used to prevent this.

Introduction

Acute appendicitis (AA) is the most frequent acute abdomen disease with the lifetime incidence of approximately 7%, which requires operative treatment due to its life threatening complications. Imitation of the symptoms makes appendicitis similar to other urgent abdomen pathologies and causes diagnostic difficulties, especially in early disease manifestation [1]. Clinical examination, laboratory and radiological tests are essentials to suspect and obtain correct diagnosis [2, 3]. Inaccurate diagnosis of appendicitis follows the unnecessary operation. A term negative appendectomy implies appendectomy for not inflamed appendicitis and reports its rates up to 40% [3, 4]. Despite the fact that surgical appendectomy is the gold standard of AA, there are studies of conservative antibiotic treatment with low morbidity and mortality rate, and a recurrence rate between 7-15 % [5]. On the other hand, delay of operative treatment can stimulate the development of complications [6-8]. Co-morbidity and negative appendectomy are strongly related with higher mortality, where co-morbidity, diagnostic failure, and the anesthetic or surgical trauma are the important factors [9]. Abdominal ultrasound and computed tomography (CT) imaging with the clinical manifestation should be the key to reduce negative appendectomy rate, but none of these completely prevents a negative operation. Moreover, the surgeon mostly relies on his clinical examination and experience, supplementing it with a scoring system [10].

The aim of our study was to identify factors associated with negative appendectomy and to propose solutions to decrease the negative appendectomy rate.

Methods

A retrospective study was performed. All adult patients (18 years and older) who underwent operation for suspected AA from 2008 to 2013 at Vilnius University Hospital Santariskiu Klinikos were included in the study. Case histories of patients, who were treated for suspected AA, were retrospectively reviewed. Data on history, clinical examination, laboratory, imaging investigation and Alvarado score were collected retrospectively from patient charts.

The following parameters were included: demographics (age, gender, body mass index), hours from onset of symptoms to diagnosis, hours from diagnosis to operation, laboratory tests (hemoglobin (Hgb), C-reactive peptide (CRP) concentration, hematocrit (Hct), white blood cell (WBC) count), Alvarado score, abdominal ultrasound or CT findings, gynecological examination results, method of operation, postoperative and pathological diagnoses.

Patients were divided into two groups based on intraoperative findings: patients in group A underwent an operation, where appendix was found to be normal (non-inflamed); patients in group B underwent appendectomy for inflamed appendix. Groups were compared for preoperative data, investigations, treatment results and pathology findings.

Categorical variables are expressed as frequencies and percentages and continuous variables as mean and standard deviation (±SD). The level of statistical signif icance was set at p < 0.05. All differences for which the probability value was less than 0.05 were considered to be significant in univariate analysis. All significant univariate risk factors were included in a multiple logistic regression model in order to identify independent predictive factors for AA. We conducted all statistical analyses using SPSS version 13.0 for Windows (SPSS Inc.).

The study was approved by regional ethics committee. Informed consent has been obtained from all individuals included in this study

Results

Five hundred and fifty four patients were included in the study: one hundred twenty seven (22.9%) patients in group A and 427 (77.1%) patients in group B. Demographics, clinical characteristics, preoperative laboratory tests results and Alvarado score distribution frequencies are summarized in Table 1. The mean age of the study population was 31 years (range 18-88). The male/female ratio was 1.23/1 in both groups. Preoperative laboratory tests results of Hgb, Hct concentrations and WBC count were significantly higher in group B (p < 0.001). Comparing Alvarado score results, there were more patients with Alvarado score less than 7 in group A with significant difference. The patients’ situations in group B were vice versa – there were more patients with Alvarado score 7 or higher. The three main Alvarado characteristics were migratory pain to right lower quadrant, tenderness in right lower quadrant and leukocytosis (Table 1).

Table 1

Demographics, clinical characteristics, preoperative laboratory tests results and Alvarado score distribution frequencies overall and between groups

	Overall (n=554)	Group A (n=127)	Group B (n=427)	p value
Age, years	31 (18-88)	30 (18-88)	32 (18-85)	>0.05
Gender
male, n (%)	306 (55)	44 (35)	262 (61)	<0.001
female, n (%)	248 (45)	83 (65)	165 (39)	<0.001
BMI (kg/m2)	25.2 ± 4.44	24.2 ± 4.84	25.51 ± 4.27	0.016
Hours
from onset of symptoms to diagnosis	24 (3-336)	24 (4-336)	20 (3-336)	<0.0001
from diagnosis to operation	2:15	3:20	1:45	>0.05
Hospital stay, days	3 (1-34)	3 (1-34)	3 (1-24)	>0.05
Hgb (g/l)	140.7 ± 16	136.6 ± 16.7	141.9 ± 15.6	<0.001
Hct (l/l)	0.41 ± 0.043	0.40 ± 0.042	0.413 ± 0.042	<0.001
WBC (*10e9/l)	13.35 ± 4.11	11.43 ± 3.78	13.91 ± 4.04	<0.0001
CRP (mg/l)	27.6 (0.2-364)	34.6 (0.2-251)	26.8 (0.2-364)	>0.05
Alvarado score < 7 (n, %)	216 (39)	75 (59)	144 (33.7)	<0.0001
Alvarado score ≥ 7 (n, %)	338 (61)	52 (41)	283 (66.3)	<0.0001
Mean of Alvarado score	6.7 ± 1.9	5.8 ± 1.9	6.9 ± 1.8	<0.0001
Migration of pain to rlq* (n, %)	537 (96.9)	119 (93.7)	418 (97.9)	0.034
Anorexia (n, %)	327 (59)	61 (48)	267 (62.5)	0.005
Nausea and vomiting (n, %)	332 (59.9)	65 (51.2)	269 (63)	0.016
Tenderness in rlq (n, %)	527 (95.1)	120 (94.5)	410 (96)	>0.05
Rebound tenderness (n, %)	299 (54)	59 (46.5)	243 (56.9)	0.047
Elevated body temperature (n, %)	60 (10.8)	17 (13.4)	44 (10.3)	>0.05
Leukocytosis (n, %)	447 (80.7)	82 (64.6)	365 (85.5)	<0.0001
Neutrophilia (n, %)	416 (75)	76 (59.8)	340 (79.6)	0.001

– right lower quadrant

– right lower quadrant

Preoperative investigations are presented in Table 2. Abdominal or pelvic ultrasonography was performed for 75% of patients and abdominal CT was performed for only 3.4% of patients.

Table 2

Percentages (%) of laboratory tests, investigations and methods of operation overall and between groups

	Overall	Group A	Group B
Laboratory tests
Hgb	100	100	100
Hct	100	100	100
WBC	100	100	100
CRP	70.2	74.8	68.9
Urine analysis	52.3	49	53
Ultrasonography	75	78	74
Visualized appendix	26.7	17	29.5
Visualized inflamed appendix	24.9	16.5	27.4
Infiltrated rlq*	22.2	20	23
Enlarged lymphnodes	6.7	11	5.4
Computed tomography	3.4	5	3
Gynecological examination	83.5	83	84
Method of operation
Open app	59.7	32	68
Laparoscopic app	24	24	24
Diagnostic laparoscopy	8.5	37	0
Conversion**	6.5	4	7
Laparotomy	1.3	2	1

– right lower quadrant

– conversion from laparoscopic to open appendectomy

24% patients underwent laparoscopy, of the 6.5% were converted to open appendectomy, 0.9% – to midline laparotomy. 59.7% patients underwent open appendec tomy, 1.3% patients were converted to midline laparotomy. Intraoperative findings are presented in Table 3. Two hundred eighty one (66%) patient of group B had phlegmonous appendicitis, 81 (19%) patient – gangrenous appendicitis, and 65 (15%) patients were with perforated gangrenous appendicitis.

Table 3

Intraoperative diagnoses and appendectomies performed

Intraoperative diagnoses	n (% of all operations)	Appendectomies performed n (% of all operations)
Macroscopically inflamed appendix
Phlegmonous appendicitis	281 (50.72)	281 (100)
Gangrenous appendicitis	81 (14.62)	81 (100)
Perforated gangrenous appendicitis	65 (11.73)	65 (100)
Macroscopically normal appendix	53 (9.57)	52 (98.1)
Adenocarcinoma of the appendix	1 (0.18)	1 (1)
Mesenteric lymphadenitis	26 (4.69)	4 (15.4)
Acute enteritis	9 (1.62)	1 (11.1)
Pelvic inflammatory disease	9 (1.62)	---
Abdominal colic	7 (1.26)	---
Terminal ileitis	2 (0.36)	---
Acute colitis	2 (0.36)	---
Acute enterocolitis	1 (0.18)	---
Ovarian cyst	1 (0.18)	---
Acute pancreatitis	1 (0.18)	---
Menstruation	1 (0.18)	---
Rupture of ovary	8 (1.44)	---
Torsion and necrosis of part of the greater omentum	4 (0.72)	---
Abscess of the greater omentum	1 (0.18)	---
Torsion and necrosis of appendix epiploica	1 (0.18)	---

Table 4 presents comparison of intra-operative findings with the pathological results. We identified 4 cases of phlegmonous appendicitis (1.4%), which were reported as non-inflamed appendix on pathologic examination. We also found 8 cases of macroscopically normally-appearing appendix, where pathology identified acute inflammation (15.4%).

Table 4

Intraoperative and pathological diagnoses of appendix

Intraoperative diagnoses	Pathological diagnoses
	n (%)	Non-inflamed appendix, n (%*)	Inflamed appendix, n (%*)
Phlegmonous appendicitis	281 (57.9)	4 (1.4)	277 (98.6)
Gangrenous appendicitis	81 (16.7)	---	81 (100)
Perforated gangrenous appendicitis	65 (13.4)	---	65 (100)
Macroscopically normal appendix	52 (10.7)	44 (84.6)	8 (15.4)
Adenocarcinoma of the appendix	1 (0.2)	1 (100)	---
Mesenteric lymphadenitis	4 (0.8)	4 (100)	---
Acute enteritis	1 (0.2)	1 (100)	---

– percentages (%) of intraoperative diagnosis

– percentages (%) of intraoperative diagnosis

All significant univariate risk factors were included in a multiple logistic regression model in order to identify independent risk factors (Table 5). Only four independent risk factors – gender, time in hours from the onset of symptoms to diagnosis, WBC count and neutrophilia were predictive of AA. R square coefficient for this model was only 0.24.

Table 5

Multivariate analysis of diagnostic criteria for acute appendicitis

Variable	p value	Odds ratio	95% CI
Gender	<0.001	0.256	0.147–0.446
Hours*	0.013	0.994	0.989–0.999
WBC	0.003	1.131	1.042–1.227
Neutrophilia	0.044	0.529	0.284–0.983

– from onset of symptoms to diagnosis

The outcomes in patients undergoing negative appendectomy or unnecessary operation did not differ from those, which had AA (Table 1) – the hospital stay was the same and there was no mortality.

Discussion

A retrospective study found the rate of negative appendectomy for 22.9%. A meta-analysis by Andersson has demonstrated that all clinical and laboratory variables are weak factors alone, but they can improve sensitivity in combination [11]. However, in our relatively large series we could find only four independent risk factors, which, if all present, could account for only 24% of cases.

The key point of proper patient care lies in a balance between the perforated appendicitis and the negative appendectomy. Delayed diagnosis is one of the main causes of perforated appendicitis. Physicians should be cautious of delaying surgery of AA since after 36 hours of untreated symptoms, the risk of perforation is increased to 5% every 12 hour period [12]. In our study, the time from the patients’ arrival to emergency room and diagnosis verification, transportation to operating theater differed twice between groups. Looking at this point, abdominal CT scan can reduce the time to diagnosis, which allows having correct diagnosis and decreases the chance of negative appendectomies [13].

Studies showed that abdominal CT scans are relatively accurate and increase correct AA diagnosis up to 95%. Moreover, there is a possibility to increase correct diagnosis to 98% when abdominal CT scan is combined with physical examination [14]. Stroman et al. reported the negative appendectomy rate close to 15% with only abdominal CT scan results and without taking into consideration the patient’s clinical picture [13]. As we see, abdominal CT scan can improve correct diagnosis, but there is still a need of physical examination. However, in the present study abdominal CT scan was only used for a small number of patients due to its costs. The abdominal ultrasound was used in most cases – 75%. The results indicated that inflamed appendix was seen in 16.5% of patients in group A. In our opinion, this has increased the number of negative appendectomies.

The Alvarado score and clinical diagnosis of appendicitis have remained relevant concepts [15]. Patients, who scored 3-6 points, with reference to Alvarado score, were more likely to have negative appendectomy. Although Alvarado score of less than 7 has been suggested to exclude AA, in our study there were 66.3% patients with inflamed appendix, who had Alvarado score of up to 7 [16]. Therefore, we would advise additional tests prior to operation, such as abdominal CT scan.

With regards to operation type, the study results have showed that laparoscopic operation was chosen more often in group A (61.41% vs. 24%). However, there was no difference in operating time or rate of complications. As for demographic data, there was difference comparing age or body mass index. Nevertheless, negative appendectomies were performed for females more often than males (65% vs. 35%, respectively).

This review has demonstrated that elements of the disease history, clinical findings and results of laboratory tests are weak individual discriminators of appendicitis.

Conclusions

In summary, AA is still often misdiagnosed and the rate of negative appendectomies remains rather high. A high rate of negative appendectomy is caused by using solely clinical examinations to diagnose AA. Additional investigations such as observation and abdominal CT should be used to prevent this.

Conflict of interest

Authors state no conflict of interest.