Keywords
appendicitis - clinical prediction rule - Alvarado - appendicitis inflammatory response
score
With a lifetime incidence of 7%, acute appendicitis is a common clinical condition,
one of the many differential diagnoses of a patient presenting with acute right iliac
fossa pain.[1] Many of the patients presenting to surgical emergency with acute right iliac fossa
pain, have alternative pathologies. A negative appendicectomy rate up to 17.5% is
noted in India following appendicectomy based on clinical suspicion.[2] This necessitates accurate diagnosis prior to performing appendicectomy. Controversies
exist regarding the management of equivocal cases, with some advocating early surgical
intervention and others opting for active observation.[3]
[4]
[5] Many clinical and imaging tools have been developed to aid in the diagnosis of acute
appendicitis. Clinical scoring systems like appendicitis inflammatory response (AIR)
score,[6] Alvarado,[7] RIPASA, pediatric appendicitis score, etc, with varying sensitivity and specificity.
Imaging modalities like ultrasonography and computer tomography are also used to evaluate
for inflamed appendix.[8]
Even though the Alvarado score is the most extensively evaluated in validation studies,
it has some drawbacks.[9]
[10] It was developed by retrospective analysis of patients who underwent appendicectomy
for suspected appendicitis and the dichotomization of variables has led to decreased
discriminative ability. AIR score was constructed from a prospective cohort, with
the grading of variables for greater discrimination, lesser reliance on symptoms,
and more on clinical signs.[7] It also includes C-reactive protein (CRP) as a component with graded scoring. CRP
is found to be well correlated with acute appendicitis in many studies.[11]
[12] AIR score assigned patients to a high probability zone with substantially higher
specificity (97%) and positive predictive value (88%) than the Alvarado score (76
and 65%, respectively).[13]
The use of scoring systems is imperative in a resource-limited setting like India.
It is useful in the clinical diagnosis, early referral from primary health care facilities,
and for deciding on surgical intervention especially when imaging is inconclusive
or not available. The study objective was to compare the performance of appendicitis
inflammatory response score and Alvarado score among patients who underwent appendicectomy
for suspected acute appendicitis.
Methods
This cross-sectional observational study was performed in the Department of General
Surgery of Government Medical College, Kottayam from January 2017 to June 2018 after
obtaining ethical clearance from the Institutional Review Board. Based on sample size
calculation the minimum sample size was estimated to be 122. Consecutive patients
aged 12 to 60 years, who were referred to the General Surgical team on-call with acute
right iliac fossa pain and were operated on for suspected acute appendicitis during
the study period and gave consent for the study were included. Patients without documented
signs, symptoms, and biochemical parameters required for calculating AIR score and
Alvarado score, those without histopathology report of the resected specimen, and
those with chronic abdominal pain were excluded from the study. The decision to operate
was taken by a senior surgical staff member. Variables that were recorded included
presence of nausea, vomiting, anorexia, migration of pain to the right lower quadrant
(RLQ), pain in the RLQ, tenderness in RLQ, rebound tenderness or muscular defense,
body temperature, white blood cell (WBC) count, proportion of polymorphonuclear leukocytes,
and CRP. Alvarado score included migration of pain (score—1), anorexia (score—1),
nausea or vomiting (score—1), tenderness in RLQ (score—2), rebound pain (score—1),
elevated temperature (score—1), leukocytosis (score—2), and shift to left (score—1).[7] AIR score included vomiting (score—1), pain in RLQ (score—1), rebound tenderness
or muscular defense (light—score 1; medium—score 2; strong—score 3), temperature (>38.5°C—score
1), polymorphonuclear leukocytes (70–80%: score 1; ≥85%: score 2), WBC count (10.0–14.9 × 103/mcl: score 1; ≥15.0 × 103/mcl: score 2), and CRP (1–4.9 mg/L: score 1; ≥5 mg/L: score 2).[6] The diagnosis was confirmed by histopathology in all resected specimens. Appendicitis
was pathologically diagnosed when infiltration of the muscularis propria by neutrophil
granulocytes was seen. For the purpose of the study, all histologically confirmed
cases were considered to have appendicitis. Statistical analysis was performed using
SPSS statistical software 23 (SPSS Inc, Chicago, IL). Continuous variables are described
as mean ( ± SD). Comparison of continuous variables was done by independent-samples
t-tests. Diagnostic accuracy was analyzed using receiver operating characteristic (ROC)
curves. Statistical significance was attributed at the 5% level.
Results
From January 2017 to June 2018, 130 patients (77 males and 53 females) who underwent
appendicectomy for suspected acute appendicitis and met the inclusion and exclusion
criteria were included in the study. The mean age was 27.5 ± 13.3 (range 12–70). 14
patients (six females and eight males) had normal appendix in histology giving a negative
appendicectomy rate of 10.7%. The mean Alvarado score among patients with acute appendicitis
in histology was 7.4 ± 1.42 as compared with 5.6 ± 1.08 among patients with normal
appendix (p < 0.001). The AIR scores were 7.3 ± 2.23 and 4.21 ± 0.97 (p< 0.001), respectively. Thirteen patients (10.3%) had a perforated appendix at surgery.
The mean Alvarado score was 7.25 for nonperforated appendix and 8.46 for perforated
appendix (p = 0.004). The mean AIR score for nonperforated cases was 6.99 and 10.23 for perforated
cases (p <0.001.)
The area under the ROC curve was 0.901 for the AIR score as compared with 0.821 for
the Alvarado score ([Fig. 1]). The Alvarado score has a sensitivity of 72% and a specificity of 79% at a score
≥6. For score ≥7, the sensitivity dropped to 46% whereas the specificity reached 93%.
The AIR score showed a sensitivity of 98% for scores ≥5 with a specificity of 36%.
The score ≥6 AIR score had a sensitivity of 77% with a specificity of 97%.
Fig. 1 The combined ROC curves for Alvarado score and AIR score. The area under ROC for
AIR score is 0.901 and the area under ROC for Alvarado score is 0.821. AIR score outperforms
Alvarado score in predicting acute appendicitis. AIR, appendicitis inflammatory response;
ROC, receiver operating characteristic.
The ROC curves for the individual parameters are shown in [Fig. 2]. Vomiting alone has a better area under ROC than nausea and vomiting combined (0.73
vs. 0.53). CRP value was the best performing with an area under ROC of 0.789, followed
by WBC count with an area under ROC of 0.762. The poorest performing among symptoms
and signs was anorexia with an area under ROC of 0.37 and the poorest performing laboratory
investigation was shifted to left with an area under ROC of 0.65. CRP of ≥10.5 had
a sensitivity of 74.1% and specificity of 64.3%, whereas a CRP ≥11.5 had a sensitivity
of 64.7% and a specificity of 71.4%.
Fig. 2 The ROC curves for individual parameters. The areas under ROC are—vomiting (0.73),
nausea or vomiting (0.533), anorexia (0.373), pain in right iliac fossa (0.50), migration
of pain (0.642), rebound tenderness (0.744), temperature (0.75), shift to left (0.650),
percentage of polymorphs (0.589), white blood cell count (0.762), and C-reactive protein
(0.789). The best performing among the individual parameters was the CRP and the worst
performing was anorexia. CRP, C-reactive protein; ROC, receiver operating characteristic.
Discussion
Appendicectomy is often the first major surgery to be undertaken by a surgeon in training.
The clinical findings supplemented by investigations help in confirming the clinical
suspicion. The use of radiological imaging has resulted in a decrease in the rates
of negative appendicectomies. However, the imaging facilities are not accessible to
all, especially in rural areas and, also during the night hours—when radiology facilities
are not available in many hospitals. Also, imaging can be inconclusive at times. This
makes the physician rely on the various clinical prediction scores.
Flum et al analyzed the data from the Washington State Database and identified 63,707
patients who underwent appendicectomy and they have noted a negative appendicectomy
rate of 15.5%.[14] Sharma et al noted a negative appendicectomy rate of 23.72%, which was 13.43% in
males and 37.25% in females.[2] Güller et al in an analysis based on the prospective database of the Swiss Association
of Laparoscopic and Thoracoscopic Surgery (SALTS) which included patients aged 12
years and over undergoing emergency laparoscopic appendicectomy between 1995 and 2006
noted a negative appendicectomy rate of 6.5% and a perforation rate of 16.5%.[15] In our study of 130 patients, the negative appendicectomy rate was noted to be 10.7%.
The mean age of the patients who had negative appendicectomy was 19.9. This was in
contrast to the mean age of patients with appendicitis in histopathology which was
28.4 (p = 0.005). This indicates the lower threshold in operating on young individuals as
compared with the elderly.
AIR score was developed by Andersson and Andersson using the prospective data of 545
patients with suspected acute appendicitis. Comparing the AIR score and Alvarado score,
the authors reported an area under ROC of 0.97 versus 0.92 (p = 0.0027) for patients with advanced appendicitis and 0.93 versus 0.88 (p = 0.0007) for all patients with appendicitis. Sixty-three percent of the patients
were classified into the low- or high-probability group with an accuracy of 97.2%,
leaving 37% in intermediate probability for further investigation. Seventy-three percent
of the nonappendicitis patients were correctly classified into low probability. Sixty-seven
percent of the advanced appendicitis patients and 37% of all appendicitis patients
were correctly classified into the high-probability zone.[6] De Castro et al among 941 consecutive patients with suspected acute appendicitis
noted that the area under the ROC curve of the AIR score was significantly better
than that for the Alvarado score (0.96 vs. 0.82, p < 0.05). The AIR score also outperformed the Alvarado score in more difficult patients,
including women, children, and the elderly.[16]
Kollár et al evaluated the AIR score and compared its performance in predicting risk
of appendicitis to both the Alvarado score and the clinical impression of a senior
surgeon. The three methods of assessment stratified similar proportions (approximately
40%) of patients to a low probability of appendicitis (p = 0.233). The false-negative rate was <8% and did not differ between the AIR score,
Alvarado score, or clinical assessment. The Alvarado score had a higher specificity
and positive predictive value than the Alvarado score (97 and 88% vs. 76 and 65%)
in assigning patients to high probability zone.[13]
Comparison between AIR score, Alvarado score, and Pediatric Appendicitis score was
done by Macco et al in a retrospective study on 747 children (<18 years). It was noted
that the area under the receiver-operating curve for the AIRS, Alvarado score, and
the Pediatric Appendicitis score was 0.90, 0.87, and 0.82, respectively (p < 0.05). In children with a low-risk acute appendicitis, false-negative rates of
14, 7, and 18% were seen for the AIR score, Alvarado score, and the Pediatric Appendicitis
score.[17] In a randomized trial on AIR score-based management of patients with suspected appendicitis,
among low-risk patients, the use of the AIR score-based algorithm resulted in less
imaging (19·2 vs. 34·5%; p < 0·001), fewer admissions (29·5 vs. 42·8%; p < 0·001), fewer negative explorations (1·6 vs. 3·2%; p = 0·030), and fewer operations for nonperforated appendicitis (6·8 vs. 9·7%; p = 0·034).[18]
In this study, the area under ROC for Alvarado score was 0.821 and for AIR score was
0.901. The Alvarado score showed better sensitivity whereas the AIR score showed better
specificity. The Alvarado score has a sensitivity of 72% and a specificity of 79%
at a score ≥6. When considering score ≥7, the sensitivity dropped to 46% whereas the
specificity reached 93%. The AIR score had a sensitivity of 98% for scores ≥5 with
a specificity of 36%. Score ≥6 showed a sensitivity drop to 77% with a specificity
of 97%. Scott et al noted a high sensitivity (90%) for AIR score of 5 or more (intermediate
and high risk) among all severities of acute appendicitis which increased to 98% for
advanced appendicitis. Ninety-seven percent specificity was noted for a score of 9
or more (high risk) with 70% among them having perforation or gangrene.[19]
In a meta-analysis by Frountzas et al RIPASA score was found to have an area under
the curve of 0.94 as compared with the Alvarado score with 0.79.[20] Chisthi et al noted that the Modified Alvarado score had an area under the curve
of 0.72 as compared with 0.94 for the AIR score and 0.91 for the RIPASA score.[21] In their study of 107 prospective patients, the AIR score outperformed the modified
Alvarado and the RIPASA scoring systems.
In the present study, it was seen that vomiting alone had a better area under ROC
than nausea and vomiting combined. This shows the lack of sensitivity of the subjective
symptom of nausea and the better discriminating ability of the objective parameter—vomiting.
Anorexia—another subjective parameter—also had poor performance. Migratory pain, rebound
tenderness, WBC count, and CRP value, all had an area under ROC above 0.7. Considering
individual parameters, the CRP value was the best performing with an area under ROC
of 0.789, followed by WBC count with an area under ROC of 0.762. The poorest performing
among symptoms and signs was anorexia with an area under ROC of 0.37. The poorest
performing laboratory investigation was shift to left with an area under ROC of 0.65.
CRP was the best overall performing individual parameter in AIRS. WBC count was the
best performing individual parameter in the Alvarado score. However, the combined
scores of Alvarado and AIR score are better than individual parameters in suspected
acute appendicitis.
This study has some limitations. First it is a relatively small size of the study
population. Second is the cross-sectional nature of the study. Third, limitation is
that only patients who underwent appendicectomy for suspected acute appendicitis were
included. Fourth is the observational nature of the study. A prospective trial including
patients with suspected acute appendicitis, evaluating the treatment decisions based
on AIR score is required in the future. Despite these limitations, it is seen from
this study and available literature that the AIR score outperforms the Alvarado score
as a clinical prediction rule in patients with suspected acute appendicitis with higher
specificity. The score is also seen to correlate with the severity of acute appendicitis.
Use of AIR score may thus decrease the rate of negative appendicectomies. The AIR
score is especially useful in situations where imaging is inconclusive and also in
resource-limited areas where radiological imaging is not available.
Conclusion
The AIR score which is a recently developed score outperforms the Alvarado score.
The ease of application of the score, the sound construction, grading of parameters,
avoiding of subjective parameters, and the introduction of CRP into scoring all make
the appendicitis inflammatory response score into an attractive clinical prediction
rule with higher specificity. The use of AIR score in resource-limited settings can
decrease the rate of negative appendicectomy.
