Meta-analysis

Systematic review and meta-analysis of the diagnostic

accuracy of procalcitonin, C-reactive protein and white
blood cell count for suspected acute appendicitis
C.-W. Yu1 , L.-I. Juan2 , M.-H. Wu3 , C.-J. Shen1 , J.-Y. Wu1 and C.-C. Lee4,5
1
Department of Emergency Medicine, Chang Gung Memorial Hospital, Keelung, and Chang Gung University College of Medicine, Taoyuan,
2
Department of Radiology, Buddhist Tzu Chi General Hospital Taipei Branch, Taipei, 3 Department of Emergency Medicine, Chang Gung Memorial
Hospital, Taoyuan, and Chang Gung University College of Medicine, Taoyuan, and 4 Department of Emergency Medicine, National Taiwan University
Hospital Yunlin Branch, Douliou, Taiwan, and 5 Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts, USA
Correspondence to: Dr C.-C. Lee, Department of Emergency Medicine, National Taiwan University Hospital Yunlin Branch, No. 579 Section 2 Yunlin
Road, Douliou City, Yunlin County 640, Taiwan (e-mail: cclee100@gmail.com)

Background: The aim was to evaluate the diagnostic value of procalcitonin, C-reactive protein (CRP)
and white blood cell count (WBC) in uncomplicated or complicated appendicitis by means of a systematic
review and meta-analysis.
Methods: The Embase, MEDLINE and Cochrane databases were searched, along with reference lists of
relevant articles, without language restriction, to September 2012. Original studies were selected that
reported the performance of procalcitonin alone or in combination with CRP or WBC in diagnosing
appendicitis. Test performance characteristics were summarized using hierarchical summary receiver
operating characteristic (ROC) curves and bivariable random-effects models.
Results: Seven qualifying studies (1011 suspected cases, 636 confirmed) from seven countries were
identified. Bivariable pooled sensitivity and specificity were 33 (95 per cent confidence interval (c.i.) 21
to 47) and 89 (78 to 95) per cent respectively for procalcitonin, 57 (39 to 73) and 87 (58 to 97) per cent
for CRP, and 62 (47 to 74) and 75 (55 to 89) per cent for WBC. ROC curve analysis showed that CRP
had the highest accuracy (area under ROC curve 0·75, 95 per cent c.i. 0·71 to 0·78), followed by WBC
(0·72, 0·68 to 0·76) and procalcitonin (0·65, 0·61 to 0·69). Procalcitonin was found to be more accurate
in diagnosing complicated appendicitis, with a pooled sensitivity of 62 (33 to 84) per cent and specificity
of 94 (90 to 96) per cent.
Conclusion: Procalcitonin has little value in diagnosing acute appendicitis, with lower diagnostic accuracy
than CRP and WBC. However, procalcitonin has greater diagnostic value in identifying complicated
appendicitis. Given the imperfect accuracy of these three variables, new markers for improving medical
decision-making in patients with suspected appendicitis are highly desirable.

Paper accepted 11 October 2012

Published online 30 November 2012 in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.9008

Introduction as 10 per cent according to recent series2 – 4 . Conditions

Acute appendicitis (AA) is the most common abdominal
surgical emergency, affecting roughly one in ten people
during their lifetime. In 1997, appendicitis-related hospital
admissions accounted for 0·6 per cent of all admissions in
the USA, resulting in approximately 1 million hospital days
and US $3 billion in hospital charges1 . The diagnosis of
appendicitis has been based largely on clinical grounds,
and the negative appendicectomy rate may be as high
commonly mistaken for AA include mesenteric adenitis,
constipation, gastroenteritis, and gynaecological diseases
such as torsion of an ovarian cyst or pelvic inflammatory
disease2 – 4 .
Recent advances in computed tomography (CT) and
ultrasound imaging have improved the accuracy of AA
diagnosis; however, heavy reliance on these advanced
imaging modalities has some disadvantages5 – 7 . First, CT is
associated with exposure to radiation and may increase the

 2012 British Journal of Surgery Society Ltd British Journal of Surgery 2013; 100: 322–329
Published by John Wiley & Sons Ltd
Diagnostic accuracy of procalcitonin, C-reactive protein and white blood cell count for suspected acute appendicitis
lifetime risk of cancer. Second, equipment and qualified
personnel for CT or ultrasonography and interpretation of
images are not available universally in all hospitals.
Delay in the diagnosis and surgery for AA may lead to
ruptured appendicitis and systemic septic complications.
Widely available laboratory tests, such as measurement
of a specific biomarker, may provide assistance in the
diagnosis of AA in most clinical settings. Conventional
white blood cell count (WBC) is neither sensitive nor
specific in the diagnosis of AA because the WBC is
increased in 70 per cent of patients with other causes of
pain in the right lower abdominal quadrant8 . C-reactive
protein (CRP), although more specific than WBC, is
less sensitive in the early stage of AA8 – 10 . CRP may be
more sensitive in detecting appendiceal perforation and
abscess formation8 – 10 . Despite the low positive predictive
value of traditional inflammatory markers, recent work has
shown that combined use of WBC and CRP may enhance
the negative predictive value. Patients experiencing lower
abdominal pain with normal values are unlikely to have
AA11,12 .
Recently, several studies have explored the role of
procalcitonin in the diagnosis of AA13 – 18 . Procalcitonin, a
precursor of calcitonin, is secreted constitutively by the C
cells of the thyroid gland and K cells of the lung. In healthy
individuals, procalcitonin is normally undetectable (serum
concentration less than 0·05 ng/ml). When stimulated
by endotoxin or inflammatory cytokines, procalcitonin is
produced rapidly by most parenchymal tissues throughout
the body19 – 21 . Unlike CRP, procalcitonin does not respond
to sterile inflammation or viral infection22 . This distinctive
characteristic makes it a popular biomarker with wide
clinical indications, including AA.
As there are few published studies, data on the usefulness
of procalcitonin in differentiating AA from other causes of
right lower quadrant pain are hard to interpret. Therefore,
a systemic review and a meta-analysis were carried out to
summarize the current evidence on the diagnostic accuracy
of procalcitonin for AA.
Methods
Standard guidelines and methods for systematic reviews
and meta-analyses of diagnostic tests were followed23 – 25 .
Data sources and searches
Electronic searches were conducted without language
restriction in the MEDLINE, Embase and Cochrane
Library databases from inception to April 2012. An update
search was done to extend the search period up to
 2012 British Journal of Surgery Society Ltd
Published by John Wiley & Sons Ltd
323
September 2012. PubMed was searched by combining two
separate queries composed of a medical subject heading
(MeSH) and text word keywords for the diagnostic test
and target outcome. Keywords used in the search included
‘appendicitis’ and ‘procalcitonin’. The search strategy was
adapted to the Embase and Cochrane Library databases.
Additional references were sought from the bibliographies
of the selected articles and other recent reviews.
Inclusion and exclusion criteria
Two reviewers independently identified articles eligible for
in-depth examination by applying the following inclusion
criteria: evaluation of procalcitonin alone or compared with
other laboratory markers such as CRP to diagnose AA or
associated complications, and sufficient data to construct a
2 × 2 contingency table. Case reports, case series, review
articles, editorials and clinical guidelines were excluded.
Two authors independently assessed all titles and abstracts
to ensure that the inclusion criteria were met. If either
of the authors considered the abstract suitable, then both
authors independently assessed the full text for suitability
for inclusion. Any discrepancies between reviewers about
articles meriting inclusion were resolved by a consensus
among three authors.
Quality assessment
The quality of the selected studies was assessed by means
of the Quality Assessment of Diagnostic Accuracy Studies
(QUADAS) criteria26 .
Data synthesis and analysis
The sensitivity and specificity were calculated for each
included data set. Owing to the negative correlation
between sensitivity and specificity caused by the different
cut-off values used in different studies, a bivariable
model was used to estimate the pooled sensitivity and
specificity of procalcitonin, CRP and WBC. The bivariable
approach assumes a bivariable distribution for logit-
transformed sensitivity and specificity values27 . Besides
accounting for study size, the bivariable model estimates
and adjusts for the negative correlation between the
sensitivity and specificity of the index test that may arise
from use of different thresholds in different studies27 .
For comparison of the discriminatory capability of
procalcitonin, CRP and WBC in different subgroups,
a hierarchical summary receiver operating characteristic
(ROC) curve was constructed. A summary ROC curve
is created by plotting the true-positive rate (sensitivity)
www.bjs.co.uk British Journal of Surgery 2013; 100: 322–329
324
against the false-positive rate (1 – specificity) at various
cut-off values from individual studies. Calculation of the
area under the ROC curve (AUROC) provides a tool
for comparison of the discriminatory capacity of different
biomarkers with different threshold values. The summary
ROC curve differs from the ROC curve in primary
diagnostic research, as each study provides one value for
sensitivity and one for specificity. A hierarchical summary
ROC curve further uses hierarchical modelling to account
for the correlation in the data via the hierarchical structure.
To deal with zero observations in 2 × 2 contingency tables,
0·5 was added to each cell, reducing performance of the
small studies. The overall sensitivity and specificity and
their 95 per cent confidence intervals (c.i.) were calculated
on the basis of the binominal distributions of the true-
positives and true-negatives.
The inconsistency index, I 2 , was calculated to quantify
the extent of heterogeneity. I 2 represents the proportion
of between-study difference that cannot be explained by
chance variation. An I 2 value over 50 per cent was consid-
ered to represent statistically significant heterogeneity.
Statistical analyses were conducted using Stata version
11.0 (StataCorp, College Station, Texas, USA), notably
the midas and metandi commands. All statistical tests
were two-sided, and statistical significance was defined
as P < 0·050.
Results
The search yielded 23 citations. After screening the title
and abstract with a focus on the age range of the study
population, 14 studies were deemed eligible for further
full-text review. Seven of these studies (1011 suspected
cases of AA, 636 confirmed) met the inclusion criteria
(Fig. 1)13,15 – 18,28 . They included 762 patients (median
132, range 101–214) with procalcitonin measurement,
685 patients (132, 40–209) with CRP measurement and
903 patients (169, 40–214) with WBC measurement. The
prevalence of AA was 0·65, 0·66 and 0·56 in studies
examining procalcitonin, CRP and WBC respectively.
All seven studies collected data prospectively, used
the same reference standard (CT or histopathology)
for outcome verification independent of the index test,
and provided clear descriptions of the patient selection
criteria and index test. They all enrolled patients with
AA and performed blood tests on admission. Pathological
examination of the surgical specimen was used as the
reference standard for patients undergoing surgery, and
CT was used as the reference test for those who did not have
surgery. All patients underwent reference tests without
differential disease ascertainment. None of the studies
 2012 British Journal of Surgery Society Ltd
Published by John Wiley & Sons Ltd
C.-W. Yu, L.-I. Juan, M.-H. Wu, C.-J. Shen, J.-Y. Wu and C.-C. Lee
Citations identified from
MEDLINE, Embase and the
Cochrane Library n = 23
Citations excluded based on screening
titles and abstracts n = 9
Potentially relevant
articles identified for
further review n = 14
Citations excluded after full-text review n = 7
Did not report diagnostic tests of interest n = 6
Did not have sufficient data to reconstruct
2 × 2 tables n = 1
Articles included in
meta-analysis n = 7
Fig. 1 Flow chart showing selection of studies for review
provided an explanation for uninterpretable test results
or participants’ withdrawal from the study. Although no
study described whether outcome verification was done by
blinding to the index test, incorporation bias was unlikely
because interpretation of surgical specimen pathology
or CT images is unlikely to be affected by biomarker
test results. Fig. S1 provides an overall picture of the
methodological quality of the studies as evaluated by the
QUADAS tool.
Table 1 summarizes the characteristics of the seven
included studies, which comprised four paediatric13,15,16,28
and three adult14,17,18 populations. Six studies used the
histopathology of the removed appendix to confirm the
diagnosis13 – 17,28 and one used CT images or surgical
pathology findings18 . Two studies used a semiquantitative
test (PCT-Q; BRAHMS Diagnostica, Berlin, Germany)
for procalcitonin measurement17,28 ; the rest used a full
quantitative test. The cut-off value of each biomarker
and corresponding sensitivity and specificity values for
AA are shown in Table 1. Five studies examined the
role of procalcitonin in diagnosing complicated appen-
dicitis. Table 2 summarizes characteristics of the five
included studies that used procalcitonin to assess com-
plicated appendicitis13,15 – 18 .The definition of complicated
appendicitis varied slightly between studies, but generally
included appendiceal perforation, mass or accumulation of
periappendiceal fluid.
Overall diagnostic indices
The pooled sensitivity and specificity for procalcitonin,
CRP and WBC are shown in Table 3. Overall, the three
www.bjs.co.uk British Journal of Surgery 2013; 100: 322–329
Diagnostic accuracy of procalcitonin, C-reactive protein and white blood cell count for suspected acute appendicitis 325

Table 1 Characteristics of the seven included studies that used biomarkers to assess acute appendicitis

Reference Country Prevalence (n) Study population PCT testing system Cut-off Sensitivity (%) Specificity (%)

Kafetzis et al.16 (2005) Greece 0·72 (212) Paediatric LUMItest PCT 0·5 ng/ml 73·0 95·0
CRP 50 mg/l 26·0 88·0
WBC 10 × 103 /mm3 82·0 59·0
Kouame et al.13 (2005) France 0·67 (101) Paediatric LUMItest PCT 0·5 ng/ml 58·0 100
Sand et al.17 (2009) Germany 0·95 (103) Adult VIDAS PCT and PCT-Q PCT 0·5 ng/ml 38·0 94·0
CRP 50 mg/l 72·0 60·0
WBC 12·1 × 103 /mm3 14·0 100
Anielski et al.14 (2010) Poland 0·67 (132) Adult LUMItest PCT 0·21 ng/ml 57·1 63·6
CRP 59·5 mg/l 45·5 100
WBC 15·6 ×103 /mm3 33·3 100
Kwan and Nager28 (2010) USA 0·55 (209) Paediatric PCT-Q CRP 30 mg/l 70 65
WBC 12·0 × 103 /mm3 71 66
Chandel et al.15 (2011) India 0·58 (40) Paediatric LUMItest PCT 0·5 ng/ml 95·6 100
CRP 50 mg/l 73·9 100
WBC 12·0 × 103 /mm3 69·6 64·7
Wu et al.18 (2012) Taiwan 0·53 (214) Adult LUMItest PCT 0·2 ng/ml 31·0 90·0
WBC 10 × 103 /mm3 61·1 72·3

PCT, procalcitonin; CRP, C-reactive protein; WBC, white blood cell count. All PCT test kits were from Diagnostica, Berlin, Germany.

Table 2 Characteristics of the five included studies that used procalcitonin to assess complicated appendicitis

Prevalence Study Definition of PCT cut-off Sensitivity Specificity

Reference Country (n) population complicated appendicitis (ng/ml) (%) (%)

Kafetzis et al.16 (2005) Greece 0·3 (212) Paediatric Extraluminal gas, periappendiceal fluid 0·5 73 95
or disseminated intraperitoneal fluid
Kouame et al.13 (2005) France 0·54 (101) Paediatric Peritonitis with intraperitoneal purulent 0·5 58 100
collection
Sand et al.17 (2009) Germany 0·23 (103) Adult Perforated and necrotizing 0·5 38 94
appendicitis
Chandel et al.15 (2011) India 0·27 (40) Paediatric Perforated appendicitis 0·7 100 100
Wu et al.18 (2012) Taiwan 0·24 (214) Adult Appendiceal perforation, phlegmon 0·5 29 95
formation, or gangrene change

PCT, procalcitonin.

Table 3 Summary of diagnostic accuracy parameters for main and subgroup analyses

No. of Sensitivity Specificity Positive Negative Area under

studies (%) (%) likelihood ratio likelihood ratio ROC curve I2

PCT for AA13 – 18 6 33 89 3·03 0·75 0·65 34·3

(21, 47) (78, 95) (1·82, 5·05) (0·66, 0·86) (0·61, 0·69) (0·0, 75·3)
PCT for complicated AA13,15 – 18 5 62 94 9·53 0·41 0·94 71·6
(33, 84) (90, 96) (4·93, 18·40) (0·20, 0·85) (0·91, 0·96) (19·4, 0·0)
PCT for adult AA14,17,18 3 36 79 1·85 0·78 0·65 16·3
(30, 42) (71, 86) (1·31, 2·63) (0·69, 0·88) (0·40, 0·99) (0·0, 91·3)
PCT for AA with parallel comparison 4 34 89 3·20 0·74 0·65 49·4
to CRP14 – 17 (20, 51) (74, 96) (1·71, 6·10) (0·27, 1·19) (0·61, 0·69) (0·0, 83·3)
CRP for AA14 – 17,28 5 57 87 4·48 0·49 0·75 70·0
(39, 73) (58, 97) (1·17, 17·07) (0·33, 0·74) (0·71, 0·78) (23·4, 8·2)
WBC for AA14 – 18,28 6 62 75 2·50 0·51 0·72 39·7
(47, 74) (55, 89) (1·47, 4·23) (0·41, 0·63) (0·68, 0·76) (0·0, 76·1)

Values in parentheses are 95 per cent confidence intervals. ROC, receiver operating characteristic; PCT, procalcitonin; AA, acute appendicitis; CRP,
C-reactive protein; WBC, white blood cell count.

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 322–329
Published by John Wiley & Sons Ltd
326 C.-W. Yu, L.-I. Juan, M.-H. Wu, C.-J. Shen, J.-Y. Wu and C.-C. Lee

1·0 1·0

0·8 0·8

0·6 0·6

Sensitivity
Sensitivity

0·4 0·4

0·2 0·2

0 0
1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0
Specificity Specificity

a Procalcitonin b C-reactive protein

1·0

0·8

0·6
Study estimate Summary point
Sensitivity

HSROC curve 95% confidence

95% prediction region
region
0·4

0·2

0
1·0 0·8 0·6 0·4 0·2 0
Specificity
c White blood cell count

Receiver operating characteristic (ROC) curve analysis for the diagnosis of patients with acute appendicitis using a procalcitonin,
Fig. 2
b C-reactive protein and c white blood cell count. The hierarchical summary ROC (HSROC) curve and bivariable mean estimate
(summary point) are shown, together with the corresponding 95 per cent confidence region and 95 per cent prediction region. The
symbol size for each study is proportional to the study size

markers showed low sensitivity but high specificity in of heterogeneity was observed for CRP and WBC. The
the diagnosis of AA. Hierarchical summary ROC curves positive likelihood ratio for the CRP test indicated that
are displayed in Fig. 2. CRP had the largest AUROC, CRP had a high diagnostic value for suspected AA. The
followed by WBC and procalcitonin. A substantial degree positive or negative likelihood ratios of procalcitonin or

 2012 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2013; 100: 322–329
Published by John Wiley & Sons Ltd
Diagnostic accuracy of procalcitonin, C-reactive protein and white blood cell count for suspected acute appendicitis
WBC were unacceptably poor for use as a stand-alone
rule-in or rule-out tool.
Subgroup analysis
Results of subgroup analysis are summarized in Table 3.
Subgroup analysis of three adult studies showed an inferior
diagnostic accuracy for procalcitonin compared with the
overall pooled estimates based on both adult and paediatric
populations. Analysis of five studies that examined the
role of procalcitonin in complicated appendicitis revealed
a greatly improved diagnostic accuracy. The high positive
likelihood ratio makes it a good diagnostic tool for rule-in
diagnosis of complicated appendicitis.
Discussion
None of the three biomarkers used for diagnosing
suspected AA was shown to be a sensitive diagnostic tool in
this study. CRP had an acceptably high positive likelihood
ratio and could be used as a rule-in diagnostic aid for
the diagnosis of AA. Despite its poor diagnostic value for
AA overall, procalcitonin had high accuracy in diagnosing
complicated AA. Patients with a positive procalcitonin
test may justifiably undergo advanced imaging studies to
exclude complicated AA before a surgical decision is made.
Future studies should examine whether a multimarker
approach or a clinical diagnostic score that incorporates
the information provided by biomarkers may improve the
accuracy of diagnosis in patients with suspected AA, and
thereby potentially improve the treatment outcome.
Although meta-analysis is not yet a widely approved
method for summarizing evidence from studies of
diagnostic tests, the authors believe that pooling the
diagnostic accuracy measures from eligible studies,
exploring the source of heterogeneity and subgroup
analysis add valuable information for both clinicians
and researchers until better studies are available. In
this meta-analysis of seven studies and 1011 patients,
CRP had the best discriminative capability of the three
laboratory markers in diagnosing AA, followed by WBC
and procalcitonin. All three markers had an unacceptably
poor sensitivity and negative likelihood ratio as a rule-
out diagnostic tool. CRP had an acceptable positive
likelihood ratio (4·48, 95 per cent c.i. 1·17 to 17·07)
and can serve as an ancillary rule-in diagnostic tool
for AA. Although not useful in diagnosing AA overall
(both complicated and uncomplicated), procalcitonin had
a high positive likelihood ratio (9·53, 4·93 to 18·40) in
identifying complicated AA and therefore represents a
valuable diagnostic aid in this setting.
 2012 British Journal of Surgery Society Ltd
Published by John Wiley & Sons Ltd
327
The superiority of CRP over procalcitonin in the
diagnosis of uncomplicated AA may be explained by the
wide infectious and non-infectious pathological spectrum
of AA15 . AA of non-infectious aetiology is caused by
obstruction of the appendiceal lumen with resulting
local inflammation29 . The identified triggers for luminal
obstruction include regional lymph node hyperplasia,
faecal stasis and faecaliths, or tumour. As regards infectious
aetiology, both bacterial and viral pathogens have been
suggested to play a role in AA13 . Regardless of the
triggering event, an uninhibited local inflammatory process
is generally believed ultimately to lead to tissue necrosis
with phlegmon, gangrene or perforation29 . Secondary
bacterial infection by an enteric pathogen may follow
the destruction of appendiceal tissue, which may help
explain the greater accuracy of procalcitonin in identifying
complicated AA over all AA.
A previous meta-analysis showed that CRP was more
accurate in the diagnosis of perforated appendicitis
(AUROC 0·87, 95 per cent c.i. 0·74 to 1·01) than all
AA (AUROC 0·75, 0·66 to 0·85)8 . The present results
suggested a better diagnostic accuracy of procalcitonin
(AUROC 0·94, 0·91 to 0·96) than CRP in diagnosing
complicated AA. The ability to discriminate between
uncomplicated and complicated AA is important because
the treatment decision is drastically different between
the two conditions. Emergency appendicectomy remains
the cornerstone of treatment for uncomplicated AA.
For complicated AA, however, an inflammatory mass
may distort the anatomy and emergency surgery is not
warranted30 .
A recent large randomized trial reconfirmed the value
of emergency appendicectomy in the management of
uncomplicated AA31 . However, a large proportion of
patients in the no-surgery group were treated successfully
with antibiotics alone, without later complications31 . This
indicated that antibiotic treatment might be an appropriate
alternative in some patients with clinically suspected
appendicitis32,33 . The procalcitonin test in combination
with clinical variables has used been successfully to guide
the clinical decision regarding use of empirical antibiotics
to treat lower respiratory tract infection or systemic
infection34,35 . It would therefore be interesting to know
whether procalcitonin could help select patients who could
safely be managed non-surgically, especially those who are
already developing complications on presentation.
Several strengths and limitations of this study must be
considered. A major strength is that likelihood ratios have
been reported in addition to sensitivity, specificity and
AUROC values. Likelihood ratios are less likely to change
with the prevalence of a disorder, and allow the clinician
www.bjs.co.uk British Journal of Surgery 2013; 100: 322–329
328
to compare different studies and calculate the post-test
probability for a target disorder36,37 .
A major limitation of the present analysis comes from
the limitations of the included studies. Five of seven studies
used the old-generation chemiluminescence procalcitonin
test (LUMItest; Brahms Diagnostica) and two even used
semiquantitative tests (PCT-Q). The LUMItest has a
functional sensitivity (20 per cent coefficient of variance) of
0·5 ng/ml. Values of less than 0·5 ng/ml, which is of great
clinical importance in appendicitis, may lack precision.
AA is a local rather than a systemic inflammation, and a
highly sensitive assay may further improve the diagnostic
performance. A more advanced procalcitonin test with
a functional sensitivity as low as 0·06 ng/ml should be
used in future studies38 – 40 . In addition, although all
included studies reported that blood tests were carried
out on admission, the interval between symptom onset
and the blood test was not reported. It is well known
that these biomarker levels fluctuate over time. Second,
significant heterogeneity was observed among studies
examining CRP and those investigating procalcitonin
in the setting of complicated appendicitis. The cut-off
values of CRP varied greatly between the studies and
the different definitions of complicated AA may be the
main reasons for this heterogeneity. Third, the study
does not provide information on the additional value of
biomarker measurements to the discrimination already
attained by clinical variables, or on whether measurement
of biomarkers may alter the clinical management of
appendicitis. Such questions need to be answered in a
well designed controlled trial. Finally, the inference that
procalcitonin may have better discriminatory capacity
for advanced appendicitis than CRP is based on the
indirect comparison assumption. The latter is based on the
homogeneous population assumption, which is unlikely to
have been met in studies from different institutions with
different case mixes. Only two studies in the present review
undertook parallel comparison of the diagnostic accuracy
of CRP and procalcitonin for complicated AA, which is
insufficient to perform a valid comparative meta-analysis.
Therefore, future studies aimed at addressing the relative
merits of procalcitonin for advanced appendicitis compared
with other inflammatory markers are needed.
Acknowledgements
J.-Y.W. and C.-C.L. contributed equally to this work. The
authors thank Pei-Shan Hsieh from Medical Wisdom,
Taipei, Taiwan, for assistance with the statistical analysis.
Disclosure: The authors declare no conflict of interest.
 2012 British Journal of Surgery Society Ltd
Published by John Wiley & Sons Ltd
C.-W. Yu, L.-I. Juan, M.-H. Wu, C.-J. Shen, J.-Y. Wu and C.-C. Lee
References
1 Davies GM, Dasbach EJ, Teutsch S. The burden of
appendicitis-related hospitalizations in the United States in
1997. Surg Infect (Larchmt) 2004; 5: 160–165.
2 Flum DR, Koepsell T. The clinical and economic correlates
of misdiagnosed appendicitis: nationwide analysis. Arch Surg
2002; 137: 799–804.
3 Flum DR, Morris A, Koepsell T, Dellinger EP. Has
misdiagnosis of appendicitis decreased over time? A
population-based analysis. JAMA 2001; 286: 1748–1753.
4 Graff L, Russell J, Seashore J, Tate J, Elwell A, Prete M et al.
False-negative and false-positive errors in abdominal pain
evaluation: failure to diagnose acute appendicitis and
unnecessary surgery. Acad Emerg Med 2000; 7: 1244–1255.
5 Doria AS, Moineddin R, Kellenberger CJ, Epelman M,
Beyene J, Schuh S et al. US or CT for diagnosis of
appendicitis in children and adults? A meta-analysis. Radiology
2006; 241: 83–94.
6 Hlibczuk V, Dattaro JA, Jin Z, Falzon L, Brown MD.
Diagnostic accuracy of noncontrast computed tomography
for appendicitis in adults: a systematic review. Ann Emerg
Med 2010; 55: 51–59.e1.
7 Obermaier R, Benz S, Asgharnia M, Kirchner R, Hopt UT.
Value of ultrasound in the diagnosis of acute appendicitis:
interesting aspects. Eur J Med Res 2003; 8: 451–456.
8 Andersson RE. Meta-analysis of the clinical and laboratory
diagnosis of appendicitis. Br J Surg 2004; 91: 28–37.
9 Wen SW, Naylor CD. Diagnostic accuracy and short-term
surgical outcomes in cases of suspected acute appendicitis.
CMAJ 1995; 152: 1617–1626.
10 Torcivia A, Vons C, Barrat C, Dufour F, Champault G.
Influence of mesh type on the quality of early outcomes after
inguinal hernia repair in ambulatory setting controlled study:
Glucamesh vs Polypropylene . Langenbecks Arch Surg 2011;
396: 173–178.
11 Sack U, Biereder B, Elouahidi T, Bauer K, Keller T,
Trobs RB. Diagnostic value of blood inflammatory markers
for detection of acute appendicitis in children. BMC Surg
2006; 6: 15.
12 Stefanutti G, Ghirardo V, Gamba P. Inflammatory markers
for acute appendicitis in children: are they helpful? J Pediatr
Surg 2007; 42: 773–776.
13 Kouame DB, Garrigue MA, Lardy H, Machet MC,
Giraudeau B, Robert M. [Is procalcitonin able to help in
pediatric appendicitis diagnosis?] Ann Chir 2005; 130:
169–174.
14 Anielski R, Kuśnierz-Cabala B, Szafraniec K. An evaluation
of the utility of additional tests in the preoperative
diagnostics of acute appendicitis. Langenbecks Arch Surg 2010;
395: 1061–1068.
15 Chandel V, Batt SH, Bhat MY, Kawoosa NU, Yousuf A,
Zargar BR. Procalcitonin as the biomarker of inflammation
in diagnosis of appendicitis in pediatric patients and
prevention of unnecessary appendectomies. Indian J Surg
2011; 73: 136–141.
www.bjs.co.uk British Journal of Surgery 2013; 100: 322–329
Diagnostic accuracy of procalcitonin, C-reactive protein and white blood cell count for suspected acute appendicitis
16 Kafetzis DA, Velissariou IM, Nikolaides P, Sklavos M,
Maktabi M, Spyridis G et al. Procalcitonin as a predictor of
severe appendicitis in children. Eur J Clin Microbiol 2005; 24:
484–487.
17 Sand M, Trullen XV, Bechara FG, Pala XF, Sand D,
Landgrafe G et al. A prospective bicenter study investigating
the diagnostic value of procalcitonin in patients with acute
appendicitis. Eur Surg Res 2009; 43: 291–297.
18 Wu JY, Chen HC, Lee SH, Chan RC, Lee CC, Chang SS.
Diagnostic role of procalcitonin in patients with suspected
appendicitis. World J Surg 2012; 11: 1744–1749.
19 Becker KL, Nylén ES, White JC, Müller B, Snider RH Jr.
Clinical review 167: Procalcitonin and the calcitonin gene
family of peptides in inflammation, infection, and sepsis: a
journey from calcitonin back to its precursors. J Clin
Endocrinol Metab 2004; 89: 1512–1525.
20 Brunkhorst FM, Heinz U, Forycki ZF. Kinetics of
procalcitonin in iatrogenic sepsis. Intensive Care Med 1998;
24: 888–889.
21 Dandona P, Nix D, Wilson MF, Aljada A, Love J, Assicot M
et al. Procalcitonin increase after endotoxin injection in
normal subjects. J Clin Endocrinol Metab 1994; 79:
1605–1608.
22 Assicot M, Gendrel D, Carsin H, Raymond J, Guilbaud J,
Bohuon C. High serum procalcitonin concentrations in
patients with sepsis and infection. Lancet 1993; 341: 515–518.
23 Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA,
Glasziou PP, Irwig LM et al. Towards complete and accurate
reporting of studies of diagnostic accuracy: the STARD
initiative. BMJ 2003; 326: 41–44.
24 Gatsonis C, Paliwal P. Meta-analysis of diagnostic and
screening test accuracy evaluations: methodologic primer.
AJR Am J Roentgenol 2006; 187: 271–281.
25 Leeflang MM, Deeks JJ, Gatsonis C, Bossuyt PM; Cochrane
Diagnostic Test Accuracy Working Group. Systematic
reviews of diagnostic test accuracy. Ann Intern Med 2008;
149: 889–897.
26 Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J.
The development of QUADAS: a tool for the quality
assessment of studies of diagnostic accuracy included in
systematic reviews. BMC Med Res Methodol 2003; 3: 25.
27 Reitsma JB, Glas AS, Rutjes AW, Scholten RJ, Bossuyt PM,
Zwinderman AH. Bivariate analysis of sensitivity and
specificity produces informative summary measures in
diagnostic reviews. J Clin Epidemiol 2005; 58: 982–990.
28 Kwan KY, Nager AL. Diagnosing pediatric appendicitis:
usefulness of laboratory markers. Am J Emerg Med 2010; 28:
1009–1015.
Supporting information
Additional supporting information may be found in the online version of this article:
Figure S1. Quality Assessment of Diagnostic Accuracy Studies (QUADAS) criteria for included studies
 2012 British Journal of Surgery Society Ltd
Published by John Wiley & Sons Ltd
329
29 Horn AE, Ufberg JW. Appendicitis, diverticulitis, and colitis.
Emerg Med Clin North Am 2011; 29: 347–368, ix.
30 Lugo JZ, Avgerinos DV, Lefkowitz AJ, Seigerman ME,
Zahir IS, Lo AY et al. Can interval appendectomy be justified
following conservative treatment of perforated acute
appendicitis? J Surg Res 2010; 164: 91–94.
31 Vons C, Barry C, Maitre S, Pautrat K, Leconte M,
Costaglioli B et al. Amoxicillin plus clavulanic acid versus
appendicectomy for treatment of acute uncomplicated
appendicitis: an open-label, non-inferiority, randomised
controlled trial. Lancet 2011; 377: 1573–1579.
32 Mason RJ, Moazzez A, Sohn H, Katkhouda N. Meta-analysis
of randomized trials comparing antibiotic therapy with
appendectomy for acute uncomplicated (no abscess or
phlegmon) appendicitis. Surg Infect (Larchmt) 2012; 13:
74–84.
33 Varadhan KK, Neal KR, Lobo DN. Safety and efficacy of
antibiotics compared with appendicectomy for treatment of
uncomplicated acute appendicitis: meta-analysis of
randomised controlled trials. BMJ 2012; 344: e2156.
34 Schuetz P, Briel M, Christ-Crain M, Stolz D, Bouadma L,
Wolff M et al. Procalcitonin to guide initiation and duration
of antibiotic treatment in acute respiratory infections: an
individual patient data meta-analysis. Clin Infect Dis 2012; 55:
651–662.
35 Schuetz P, Christ-Crain M, Müller B. Procalcitonin and
other biomarkers to improve assessment and antibiotic
stewardship in infections – hope for hype? Swiss Med Wkly
2009; 139: 318–326.
36 Akobeng AK. Understanding diagnostic tests 2: Likelihood
ratios, pre- and post-test probabilities and their use in clinical
practice. Acta Paediatr 2007; 96: 487–491.
37 Halkin A, Reichman J, Schwaber M, Paltiel O, Brezis M.
Likelihood ratios: getting diagnostic testing into perspective.
QJM 1998; 91: 247–258.
38 Lacroix L, Manzano S, Galetto A, Gervaix A. Procalcitonin
measurement for detection of serious bacterial infection in
febrile children: comparison between two automated
immunoassays. Clin Biochem 2012; 45: 593–595.
39 Manzano S, Bailey B, Girodias JB, Cousineau J, Delvin E,
Gervaix A. Comparison of procalcitonin measurement by a
semi-quantitative method and an ultra-sensitive quantitative
method in a pediatric emergency department. Clin Biochem
2009; 42: 1557–1560.
40 Sanders RJ, Schoorl M, Dekker E, Snijders D, Boersma WG,
Ten Boekel E. Evaluation of a new procalcitonin assay for
the Siemens ADVIA Centaur with the established method on
the B.R.A.H.M.S Kryptor. Clin Lab 2011; 57: 415–420.
www.bjs.co.uk British Journal of Surgery 2013; 100: 322–329