138 patients were enrolled in our study between 01.01.2019 and 01.01.2020. Average age of the patients was 32 years (18-67 years). There were 93 female and 45 male participants. The average value of the modified Alvarado score was 6.5: 1-n=3, 2-n=17, 3-n=13, 4-n=27, 5-n=16, 6-n=12, 7-n=27, 8-n=19, 9-n=3, 10-n=1. Main groups were the following: 1-4 scores (discharge) n=60, 5-6 scores (observation) n=28, 7-10 scores (emergency surgery) n=50.
The value of the modified Alvarado score in the diagnosis of acute appendicitis: the modified Alvarado scores of the patients were compared with the final histology findings (0 – acute appendicitis was not confirmed, as no surgery was performed or the histology was negative (n=87), group 1 – another disease was confirmed (tumor, diverticulum) (n= 5), group 2 – slight inflammation (simple acute appendicitis, phlegmonous acute appendicitis, superficial acute appendicitis) (n=20), group 3 – severe inflammation (ulcero-phlegmonous acute appendicitis, gangrenous acute appendicitis, perforation) (n=26). Comparing the scores with the histological findings, specificity of the modified Alvarado score was 84.78% (95% Clopper-Pearson confidence interval: 75.79% to 91.42%), sensitivity was 97.83% (95% confidence interval: 88.47% to 99.95%).
As a result, a strong, significant Spearman’s rank correlation was found between the modified Alvarado score and the final histology finding: 0.796 (p<0.001). The above figure shows that no negative pathological findings were present above a score of 4, that is the value of the pathological result was 0 (Figure 1).
The pathological finding was coded as 0 if the score was 0 or 1 (no appendicitis present) and 1 if the score was 2 or 3 (appendicitis was present), ROC analysis showed an area under the curve value of 0.968 (95% C-I: 0.939, 0.997, p<0.001), which suggests good separability (Figure 2).
In our study, we tried to determine whether the cut-off values used in case of the original Alvarado score are applicable in case of the modified Alvarado score as well: the previous two cut-off values were: score of 4 (acute appendicitis is not likely to be present below this value) and score of 7 (surgery is recommended above this value regarding the fact that the likelihood of appendicitis is high), and there was a “gray zone” of scores 5-6 in case of which further observation or additional imaging studies (CT) were required. We found that the risk of appendicitis is low in case of a score of 4 using the modified Alvarado score as well, and the risk is high above a score of 7. A score of 5-6 indicates closer observation and repeated score calculation, in some cases, additional imaging studies may be required (acute CT examination) (Table 2 and Figure 3).
Table 2: Cross tabulation of the modified Alvarado score and the histology numerical code
|
Pathology numerical code
|
|
|
0
|
1
|
2
|
3
|
Total
|
Modified Alvarado score
|
0
|
3
|
0
|
0
|
0
|
3
|
|
1
|
3
|
0
|
0
|
0
|
3
|
|
2
|
17
|
0
|
0
|
0
|
17
|
|
3
|
13
|
0
|
0
|
0
|
13
|
|
4
|
25
|
1
|
1
|
0
|
27
|
|
5
|
16
|
0
|
0
|
0
|
16
|
|
6
|
8
|
1
|
2
|
0
|
11
|
|
7
|
2
|
2
|
8
|
14
|
26
|
|
8
|
0
|
1
|
7
|
10
|
18
|
|
9
|
0
|
0
|
2
|
1
|
3
|
|
10
|
0
|
0
|
0
|
1
|
1
|
|
|
|
|
|
|
|
Total
|
|
87
|
5
|
20
|
26
|
138
|
Horizontal line presents the possible cutoff values, blue line refers to sensitivity (in function of the possible cutoff values) and red line presents specificity.
The patient was discharged based on the clinical status in 78 cases, the Alvarado score was below 4 in all of these cases. Repeated consultation was performed 24 hours later in one case, and the surgeon decided to perform surgery (repeated Alvarado score was 7). Surgery was performed in 60 patients: laparoscopic appendectomy was performed in 56 cases, adhesiolysis was performed in 1 case, right hemicolectomy was performed in 1 case, Hartmann's procedure was performed due to sigmoid diverticulitis in 1 case, and only explorative surgery was performed in 1 case. From the 56 laparoscopic appendectomies, drain was left in the patient in 23 cases. No conversion was required. Hemolock clips were used for the closure of appendicular stumps in 55 cases, Endo GIA was required in one case due to the severity of the inflammation.
In our study, 60 patients had surgery, 37 of these patients received Zinacef (cefuroxime), Klion (metronidazole) antibiotic prophylaxis once (n=37), 3 patients received (ciprofloxacin), Klion (metronidazole) prophylaxis due to penicillin allergy (n=3). Naturally, prophylactic therapy may be or have to be continued based on the intraoperative image, the severity of the inflammation, antibiotic therapy may be switched based on for example the intraoperative bacteriological sampling. In our study, in 3 cases, additional IV Rocephin (ceftriaxone) - Klion (metronidazole), in 1 case Tienam (imipenem/cilastin) - Klion (metronidazole) antibiotic therapy was continued, and 15 patients received oral antibiotic after the initial intravenous antibiotic therapy (Zinnat [cefuroxime] + Klion [metronidazole] n=13, Cifran [ciprofloxacin] - Klion [metronidazole] n=1).
In addition, we evaluated the correlation between the severity of the inflammation (based on the final histology report) and the type and duration of the antibiotic therapy. Based on the severity of the inflammation (pathological finding), we classified the patients into four groups: 0 – acute appendicitis was not confirmed, as no surgery was performed or the histology was negative (n=87), group 1 – another disease was confirmed (tumor, diverticulum) (n= 5), group 2 – slight inflammation (simple acute appendicitis, phlegmonous acute appendicitis, superficial acute appendicitis) (n=20), group 3 – severe inflammation (ulcero-phlegmonous acute appendicitis, gangrenous acute appendicitis, perforation) (n=26). Based on antibiotic therapy, 3 groups were created: 0 - no antibiotic therapy was administered (n= 80), group 1 - one shot IV antibiotic prophylaxis (n=40), group 2 - antibiotic therapy (n=18).
Significant correlation was found between the severity of the pathology and the duration of the antibiotic therapy. Spearman’s rank correlation confirmed this correlation to be significant, but moderate (r=0.605, p<0.001). The statistical correlation confirms the correctness of the clinical judgment, which means that therapeutic (longer) antibiotic therapy is administered if it is indicated.
Patients were classified into 2 groups based on the insertion of a drain: 0 - no drain was used as no surgery was performed or surgery was performed but drain was not used (n=115), group 1 – drain was inserted (n=23).
Based on the pathology, the patients were classified into 4 groups as mentioned above: 0 – acute appendicitis was not confirmed, as no surgery was performed or the histology was negative (n=87), group 1 – another disease was confirmed (tumor, diverticulum) (n= 5), group 2 – slight inflammation (simple acute appendicitis, phlegmonous acute appendicitis, superficial acute appendicitis) (n=20), group 3 – severe inflammation (ulcero-phlegmonous acute appendicitis, gangrenous acute appendicitis, perforation) (n=26).
Relationship was found between drain insertion and the severity of inflammation, Fisher's exact test found this relationship to be significant (p<0.001), which means that drain is not inserted routinely into the abdominal cavity in case of appendectomies, the decision was made by the operating surgeon depending on the intraoperative finding. The clinical decision is confirmed by the result of the statistical analysis regarding the histology finding (Table 3).
Table 3: Relationship between pathological severity and drain insertion (cross tabulation)
|
|
Drain
|
Total
|
0
|
1
|
Pathológia számkód
|
0
|
Count
|
81
|
6
|
87
|
% within Pathology number code
|
93,1%
|
6,9%
|
100,0%
|
1
|
Count
|
3
|
2
|
5
|
% within Pathology number code
|
60,0%
|
40,0%
|
100,0%
|
2
|
Count
|
14
|
6
|
20
|
% within Pathology number code
|
70,0%
|
30,0%
|
100,0%
|
3
|
Count
|
17
|
9
|
26
|
% within Pathology number code
|
65,4%
|
34,6%
|
100,0%
|
Total
|
Count
|
115
|
23
|
138
|
% within Pathology number code
|
83,3%
|
16,7%
|
100,0%
|
We analyzed the correlation between the duration of hospitalization (days) and the severity of the inflammation (based on the pathological finding: 0 – acute appendicitis was not confirmed, as no surgery was performed or the histology was negative (n=87), group 1 – another disease was confirmed (tumor, diverticulum) (n= 5), group 2 – slight inflammation (simple acute appendicitis, phlegmonous acute appendicitis, superficial acute appendicitis) (n=20), group 3 – severe inflammation (ulcero-phlegmonous acute appendicitis, gangrenous acute appendicitis, perforation) (n=26), that is whether there is correlation between the severity of the inflammation and the duration of hospitalization. In our study, 13 patients spent 24 hours in the hospital, 23 patients spent 48 hours in our Department, 9 patients were hospitalized for 3 days, 2 patients were hospitalized for 4 days, 4 patients were hospitalized for 5 days. 5 patients were hospitalized for an even longer period of time: the days of hospitalization was 6 in 4 cases and 9 in one case.
The average duration of hospitalization was 2,625 days (1-9 days).
Spearman's rank correlation found positive and significant relationship between the hospitalization time and severity, meaning that hospitalization is longer in case the inflammation is likely to be more severe. Rank correlation was 0.71 (p<0.001), moderately strong.
Ultrasound examination was performed in all cases, regarding the fact that the ultrasound examination is part of the modified Alvarado score: in 88 cases, the ultrasound did not confirm acute appendicitis, appendicitis was confirmed in 50 cases based on the ultrasound image. Additional CT examination was performed in 11 cases, when the ultrasound was negative. From these examinations, CT confirmed appendicitis in 7 cases, 3 patients had negative CT findings, 1 patient had intestinal conglomerate in the right lower abdomen with abscess formation.
The diagnostic accuracy of the imaging methods was calculated by comparing the results of the ultrasound examination with the final histology findings: sensitivity was 82.6%, specificity was 87%. In can be concluded, that the predictive value of this imaging method is good (Table 4).
Table 4: Cross tabulation of ultrasound (US) finding and histology finding
|
Histology finding
|
Total
|
|
0
|
1
|
|
US finding
|
0
|
Count
|
80
|
8
|
88
|
|
% within Histology
|
87.0%
|
17.4%
|
|
|
1
|
Count
|
12
|
38
|
50
|
|
% within Histology
|
13.0%
|
82.6%
|
|
|
Total
|
Count
|
92
|
46
|
138
|
|
|
|
Specificity and sensitivity of the ultrasound were highlighted as bold italic fonts.