Received: 13 May 2018 | Revised: 31 December 2018 | Accepted: 3 January 2019

DOI: 10.1111/1756-185X.13487

ORIGINAL ARTICLE

HLA‐B27 prevalence in axial spondyloarthritis patients and

in blood donors in a Lebanese population: Results from a
nationwide study

Nelly Ziade1,2 | Ghada Abi Karam1,2 | Georges Merheb3,4 | Iyad Mallak5 |

Laure Irani6,7 | Elie Alam8 | Jamil Messaykeh9 | Jeanine Menassa6,10 |
Kamel Mroue’11 | Imad Uthman12 | Abdel Fattah Masri12 | Pierre Ghorra13 |
Torsten Witte14 | Xenofon Baraliakos15,16

1
Rheumatology Department, Saint‐Joseph
University, Beirut, Lebanon Abstract
2
Rheumatology Department, Hotel‐Dieu de Aim: To calculate the prevalence of human leukocyte antigen (HLA)‐B27 in axial
France Hospital, Beirut, Lebanon
spondyloarthritis patients (axSpA) compared to blood donors (BD) in Lebanon, to
3
Rheumatology Department, Holy Spirit
identify the clinical and radiological findings associated with HLA‐B27 and to esti-
University, Kaslik, Lebanon
4
Rheumatology Department, Notre‐Dame mate the proportion of patients fulfilling the clinical arm of the Assessment of the
des Secours University, Lebanon Spondyloarthritis International Association (ASAS) criteria.
5
Radiology Department, Hotel‐Dieu de Method: Consecutive Lebanese adult axSpA patients fulfilling the ASAS classifica-
France, Beirut, Lebanon
6 tion criteria were included from 12 rheumatology clinics across Lebanon. BD served
Lebanese University Hospital, Beirut,
Lebanon as controls. A binary logistic regression was used to study the association between
7
Rizk University Medical Center, Beirut, HLA‐B27 and the disease features.
Lebanon
8
Results: A total of 247 individuals were included (141 axSpA patients and 106 BD).
Levant Hospital, Sin‐el‐Fil, Lebanon
9 The prevalence of HLA‐B27 was 3.8% in BD and 41.1% in axSpA. Overall, 39.7% of
Monla Hospital, Tripoli, Lebanon
10
Geitawi Hospital, Beirut, Lebanon
the axSpA patients fulfilled the clinical arm of the ASAS classification criteria.
11
Al Zahra University Hospital, Beirut, Sensitivity of HLA‐B27 for axSpA was 41.1%, specificity was 96.2%, positive predic-
Lebanon tive value was 93.6%, and negative predictive value was 55.13%. Positive likelihood
12
American University of Beirut Medical
ratio (LR) was 10.9 and negative LR was 1.63. We found a positive association of
Center, Beirut, Lebanon
13 HLA‐B27 with family history of SpA and psoriasis.
Blood Transfusion Center, Hotel‐Dieu de
France, Beirut, Lebanon Conclusion: Our study confirmed a low prevalence of HLA‐B27 in axSpA patients
14
Medical University, Hannover, Germany and BD in this Lebanese population, However, we found a high specificity and posi-
15
Rheumazentrum Ruhrgebiet, Herne, Germany tive LR, as well as the same number of axSpA patients fulfilling the clinical arm of the
16
Ruhr-Universität Bochum, Bochum, Germany
ASAS criteria as in European studies. HLA‐B27 is therefore valuable for identification
Correspondence of axSpA in Lebanese patients despite the overall low prevalence in this population.
Nelly Ziade, Assistant Professor at
Our results may guide future evaluations the role of HLA‐B27 in planning local refer-
Saint‐Joseph University, Rheumatology
Department, Consultant Rheumatologist ral strategies.
at Hotel‐Dieu de France Hospital,
Rheumatology Department, Beirut, Lebanon.
KEYWORDS
Email: [email protected]
ankylosing spondylitis, biomarkers, diagnostic tests, epidemiology, genetics
Funding information
AbbVie Deutschland GmbH & Co. KG

[Correction added on 22 February 2019, after first online publication: Affiliation “Ruhr-Universität Bochum, Bochum, Germany” has been added for author Xenofon Baraliakos.]

708 | © 2019 Asia Pacific League of Associations for wileyonlinelibrary.com/journal/apl Int J Rheum Dis. 2019;22:708–714.
Rheumatology and John Wiley & Sons Australia, Ltd
ZIADE et al. | 709

1 | I NTRO D U C TI O N of the country, reflecting the population and healthcare systems’

densities).
Axial spondyloarthritis (axSpA) is a frequent inflammatory rheumatic The inclusion criteria were the following: adult Lebanese patients
disease with a prevalence ranging from 0.1% to 2% in Caucasian popula- (patients of Lebanese origin, living in Lebanon) who were diagnosed
tions.1,2 AxSpA affects mainly young adults and can be significantly dis- with axSpA by their treating rheumatologist and then tested to fulfill
2,3
abling, with serious professional impact and high socioeconomic cost. the ASAS criteria to be eligible for the study.
The prevalence of SpA has been roughly estimated to be 0.3% in The control arm included consecutive BDs recruited at a trans-
Lebanon, according to the Community Oriented Program for Control fusion center in Hotel‐Dieu de France hospital, one of the largest in
of Rheumatic Diseases study (COPCORD).4 According to another the country, during the same study period.
nationwide study published in 1997 by Lebanese rheumatologists, Non‐Lebanese patients and BDs (such as Syrian and Palestinian
SpA accounted for 8.6% and ankylosing spondylitis (AS) accounted refugees) and Lebanese patients of Armenian ancestry were identi-
for 2.8% of rheumatology consultations using the Amor, Eureopean fied by self‐declaration of both parents’ origins and were excluded to
Spondyloarthritis Study Group or modified New York criteria.5-8 keep the study population as homogeneous as possible.
Recent evidence shows that an earlier diagnosis of SpA leads Each axSpA patient and BD gave written consent to participate
to optimizing patient management.9 The Assessment of the in the study.
Spondyloarthritis International Association (ASAS) classification cri- Clinical and laboratory assessments of the disease were per-
teria for axSpA, published in 2009,10 rely on sacroiliitis findings on formed in all axSpA patients by their treating rheumatologist at the
imaging with one additional SpA feature (imaging arm) or HLA‐B27 study visit, including basic demographic data, disease characteristics,
(human leucocyte antigen B27) positivity with two additional SpA treatment categories, disease activity measures (Bath Ankylosing
features (clinical arm), in patients with chronic low back pain and less Spondylitis Disease Activity Index [BASDAI], Ankylosing Spondylitis
than 45 years of age at onset of symptoms. Disease Activity Score [ASDAS]) and extra‐articular manifestations.
Therefore, HLA‐B27 plays one of the central roles in axSpA classifi- Hip involvement was defined according to the imaging data available
cation.11,12 It is a class I antigen of the major histocompatibility complex in the patient's file. Familial history of SpA, psoriasis and inflamma-
that is present in 6%‐8% of the general population and in 80%‐95% of tory bowel disease (IBD) as reported by a physician was investigated
AS patients in western European and North American populations.13 during the study visit.
However, the prevalence of HLA‐B27 in axSpA patients is found to be Frozen ethylenediaminetetraacetic acid (EDTA) blood of axSpA
lower in Japan and in most Arab and Middle Eastern populations, com- patients and BDs were tested for HLA‐B27 (by PCR) at the patient's
pared to western European populations, possibly due to different genetic usual laboratory. Additionally, frozen EDTA blood was sent to the
backgrounds.14-18 The lowest figures in the region were found in Lebanon central laboratory in Hannover, Germany. Genomic DNA was iso-
with a prevalence of 26.3% in AS, 13.8% in SpA and 1.4% in the general lated from 1 mL of thawed blood. The commercially available CE‐
population, according to the latest published study in 1997.15 These prev- labelled kit GenoQuick HLA‐B27 (Version 2.0; Cat No: 31196; Hain
alence rates may affect the diagnostic value of HLA‐B27 and impact the Lifescience, Nehren, Germany) was used for measuring HLA‐B27.
local application of published referral strategies, which are mainly elabo- Using this kit, all the subtypes of HLA‐B27 are detected. DNA ob-
rated in European populations with a high HLA‐B27 prevalence. tained from two HLA‐B27 positive and from two HLA‐B27 neg-
Since the last published HLA study in 1997, new methods for ative axSpA patients from Germany were used as positive and
HLA‐B27 testing were used (ie polymerase chain reaction [PCR]), negative controls. This double testing aimed at investigating the
with higher sensitivity than the previously used flow cytometry. performance quality of the local laboratories, as it was previously
Moreover, the new ASAS classification criteria now allows earlier questioned by some experts and incriminated in the low HLA‐B27
and possibly wider axSpA identification. These new elements mo- prevalence.
tivated this study on HLA‐B27 prevalence in axSpA patients and in Comparison between groups was performed using a Chi‐
the general Lebanese population, with blood donors (BDs) used as square test and Fisher's exact test for qualitative variables and
surrogates for the latter. Student's test for quantitative variables. Binary logistic regression
The primary objective of this study was to calculate the preva- was used to study the association between HLA‐B27 positivity
lence of HLA‐B27 in axSpA patients compared to BDs in Lebanon. and clinical, biologic and radiologic features (radiographic sacro-
The secondary objectives were to identify the clinical and radio- iliitis). Multivariate logistic regression included the variables that
logical factors associated with HLA‐B27 and to estimate the propor- were statistically associated in bivariate analysis in the final model.
tion of patients fulfilling the ASAS criteria clinical arm. Agreement for the laboratory results was tested using Cohen's
kappa coefficient.
Statistical significance was fixed at a P value <0.05. Statistical
2 | M ATE R I A L S A N D M E TH O DS analyses were performed using the IBM SPSS Statistics software
version 23 (IBM, Armonk, NY).
The axSpA patients were recruited from 12 rheumatology clinics The study was approved by the Saint‐Joseph University of Beirut
across Lebanon (Beirut [42%], Mount Lebanon, north and south ethics committee.
710 | ZIADE et al.

3 | R E S U LT S as onset before 16 years old; r = 0.15, P = 0.052) and to a negative

association with IBD (r = −0.14, P = 0.078). Other disease features
3.1 | Population demographics and activity scores were not associated with HLA‐B27 status; spe-
cifically, no association was found with disease activity and radio-
Data were collected for 247 individuals (141 axSpA patients and 106
graphic involvement.
consecutive BDs).
Using a binary logistic regression for multivariate analysis,
The mean age was 35.7 years (SD 13.0), axSpA patients were
HLA‐B27 remained positively associated with family history of SpA
older than the BD participants (42.0 years [SD 12.6] vs 27.2 years
(ß = 1.5, P = 0.001) and psoriasis (ß = 0.9, P = 0.041), whereas it
[SD 7.5], P < 0.001).
tended to be negatively associated with IBD (current or previous)
Males were 59.0% of the analyzed individuals (61.7% in axSpA,
(ß = 1.1, P = 0.055).
53.6% in BD, without statistically significant differences between
Most of the patients with HLA‐B27 fulfilled the clinical arm for
axSpA and BDs [P = 0.297]).
ASAS classification (56 of 58 HLA‐B27 positive patients), which was
39.7% of our total axSpA population. In these HLA‐B27 positive pa-
3.2 | HLA‐B27 diagnostic properties tients, the ASAS classification was based mostly on the inflamma-
tory nature of back pain and good non‐steroidal anti‐inflammatory
Comparisons between local HLA‐B27 (tested in Lebanon) and cen-
drug response (>50% of cases). Details of SpA features are shown
tral HLA‐B27 (tested in Hannover) were done in one‐third of the par-
in Figure 1.
ticipants (sample of 88 individuals). The agreement between local
and central testing was 96.6% (85/88 tests were identical, Cohen's
kappa coefficient was 0.930, P < 0.001).
4 | D I S CU S S I O N
The prevalence of HLA‐B27 was 3.8% in the BD (assumed
healthy) population (4/106 BDs) and 41.1% in the axSpA patients
Our study confirmed the lower prevalence of HLA‐B27 in axSpA pa-
(48/141).
tients (41.1%) and BD participants (3.8%) in Lebanon, compared to
The diagnostic properties of HLA‐B27 in axSpA were: sensi-
northern populations.14 These results are in line with the prevalence
tivity 41.1% (58/141 patients), and specificity 96.2%. The positive
found in normal populations from the Arab and Middle Eastern re-
predictive value was 93.6%, and the negative predictive value was
gion (0.3%‐6.8%) but are still lower than the published prevalences
55.1%. The positive likelihood ratio (LR+) was 10.9, and the nega-
for axSpA (42.7%‐84%).16
tive likelihood ratio (LR−) was 1.6 (diagnostic properties are shown in
However, the HLA‐B27 prevalence presented here is higher than
Contingency Table 1).
the one published in 1997 (26%), which could be due to more sen-
sitive HLA‐B27 testing methods (PCR vs flow cytometry). This low
3.3 | AxSpA population characteristics and ASAS HLA‐B27 prevalence does not seem to mirror lower axSpA preva-
SpA features lence as previously suggested,19 since the estimated disease prev-
alence is 0.3% according to the COPCORD study (which could be
Population and disease characteristics are presented in Table 2.
even higher since the COPCORD methodology was based on tele-
Using a bivariate analysis, we found a positive association of
phone contacts that may underestimate disease prevalence).
HLA‐B27 with family history of SpA (r = 0.32, P < 0.001), psoriasis
Regarding the reasons for this lower HLA‐B27 prevalence in
(r = 0.18, P = 0.029), thoracic axial symptoms, that is dorsal pain
our country, genetic and ethnic factors are likely. Lebanon encom-
and sterno‐costal pain (r = 0.22, P = 0.009) and tragus‐to‐wall dis-
passes a great mix of cultural, religious, and ethnic groups, which
tance (r = 0.198, P = 0.036). There was a tendency toward positive
have been developing for more than five millennia. A recent ge-
association with early age of symptom onset (juvenile onset defined
netic study found that present‐day Lebanese derive most of their
ancestry from a Canaanite‐related population (ancestors of the
TA B L E 1 Contingency table for HLA‐B27 in axSpA patients and Phoenicians), which implies substantial genetic continuity in the
BD participants
Levant region since at least the Bronze Age. In addition, the study
axSpA BD found Eurasian ancestry in the Lebanese not present in the Bronze

HLA‐B27+ 58 4 62 PPV Age or earlier Levantines, and estimated that this Eurasian ancestry
93.6% arrived in the Levant approximately 3750‐2170 years ago during
HLA‐B27− 83 102 185 NPV a period of successive conquests by distant populations. 20 This
55.1% Canaanite‐related origin with the later Eurasian addition makes up
141 106 247 the current genetic heritage of the Lebanese population. Religious
Sensitivity Specificity affiliation has also become a substitute in some respects for eth-
41.1% 96.2% nic affiliation. 21 The present Lebanese Constitution officially ac-
axSpA, axial spondyloarthritis; BD, blood donors; HLA, human leukocyte knowledges 18 religious groups (mostly Muslims and Christians,
antigen; NPV, negative predictive value; PPV, positive predictive value with some ethnic minorities). 22,23 However, population genetics
ZIADE et al. | 711

TA B L E 2 Characteristics of axSpA patients by HLA‐B27 status

All axSpA HLA‐B27+ HLA‐B27− P

Age, mean (SD) 42.03 (12.6) 41.03 (13.3) 42.7 (12.1) 0.431
Male gender, % 61.7 63.8 60.2 0.402
Non‐radiographic axSpA, % 34 25.9 39.8 0.062
Mean disease duration, years (SD) 6.6 (7.6) 7.9 (1.2) 5.7 (0.7) 0.114
Juvenile onset, before 16 y, % 5.7 10.3 2.4 0.052
IBP ASAS criteria, % 80.9 82.5 81.7 0.548
IBP Berlin criteria, % 76.6 81.0 73.5 0.202
IBP Calin criteria, % 90.6 91.2 90.2 0.456
Biologics, % 42.6 43.1 42.2 0.524
csDMARDs, % 29.1 27.4 30.1 0.575
Current NSAIDs, % 69.5 75.9 65.1 0.118
Good response to NSAIDs, % 60.3 63.8 57.8 0.296
Axial manifestations (according to patients’ symptoms) %
Buttock 92.2 96.6 89.2 0.095
Lumbar 95.7 94.8 96.4 0.480
Thoracic 31.2 43.1 22.9 0.009
Cervical 41.1 48.3 36.1 0.103
Number of axial locations (SD) 2.6 (0.95) 2.8 (0.13) 2.4 (0.09) 0.021
Hip involvement (according to the imaging data) % 24.8 25.9 24.1 0.481
Peripheral manifestations %
Arthritis 36.9 36.2 37.3 0.517
Enthesitis 29.8 36.2 25.3 0.114
Dactylitis 2.1 5.2 0 0.069
Extra‐articular manifestations %
Psoriasis 19.1 27.6 13.3 0.029
IBD 18.4 12.1 22.9 0.078
Uveitis 8.5 10.3 7.2 0.360
Preceding infection 1.4 1.7 1.2 0.655
Family history (as reported by a physician) %
Psoriasis 11.3 13.8 9.6 0.307
IBD 12.1 17.2 8.4 0.095
SpA 29.1 46.6 16.9 <0.001
Schober index, cm (SD) 4.4 (1.9) 4.2 (0.3) 4.5 (0.2) 0.268
Chest expansion, cm (SD) 4.7 (1.2) 4.6 (1.2) 4.8 (0.1) 0.405
Tragus‐wall, cm (SD) 12.1 (3.9) 13.0 (0.7) 11.5 (0.3) 0.036
BASDAI (SD) 3.6 (1.7) 3.8 (0.2) 3.5 (0.2) 0.327
BASDAI ≥4, % 33.3 34.5 32.5 0.810
ESR (SD) 24.1 (18.9) 24.0 (2.7) 24.1 (1.9) 0.969
CRP concentration, mg/dL (SD) 8.0 (14.0) 8.8 (2.3) 7.4 (1.2) 0.203
Elevated CRP % 36.9 43.1 32.5 0.541
ASDAS‐ESR, (SD) 2.7 (0.9) 2.7 (0.1) 2.7 (0.1) 0.371
ASDAS‐CRP (SD) 2.7 (1.1) 2.8 (0.1) 2.7 (0.1) 0.554
High and very high disease activity according to ASDAS‐CRP (>2.1), % 76.6 74.7 79.3 0.905

ASAS, Assessment of the Spondyloarthritis International Association; ASDAS, Ankylosing Spondylitis Disease Activity Score; axSpA, axial spondyloar-
thritis; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; CRP, C‐reactive protein; csDMARDs, conventional synthetic disease‐modifying
anti‐rheumatic drugs; ESR, erythrocyte sedimentation rate; HLA, human leukocyte antigen; IBD, inflammatory bowel disease; IBP, inflammatory back
pain; NSAIDs, non‐steroidal anti‐inflammatory drugs.
Statistically significant associations are marked in bold.
712 | ZIADE et al.

F I G U R E 1 Prevalence of Assessment of the Spondyloarthritis International Association spondyloarthritis (SpA) classification features
in axial SpA patients by human leukocyte antigen (HLA)‐B27 status. CRP, C‐reactive protein; IBD, inflammatory bowel disease; IBP,
inflammatory back pain; NSAIDs, non‐steroidal anti‐inflammatory drugs

experts have noted throughout the last decades that there are no found a positive association with uveitis18 and reactive arthritis, 31
24,25
significant genetic differences between religious groups. In which was not found in the present study. An association with high
particular, a genetic study showed that HLA gene frequencies for disease activity has been suggested in previous studies, when a
loci A and B are shown to cross Muslim‐Christian lines but gener- positive correlation with high CRP was suggested, but this finding
ate a distinguishably different profile for the Armenian immigrant was not confirmed in later studies. 27 In our study, we found no as-
25
subpopulation. sociation with CRP, BASDAI or ASDAS. Moreover, we did not find
Despite the low prevalence, HLA‐B27 was found to have high an association with radiographic damage, as reflected by the lack
specificity and positive LR similar to what was previously published of association with the radiographic axSpA form, nor with hip in-
from European studies,10 which makes this test highly valuable for volvement, as was suggested in previous studies. 32 Previous stud-
axSpA identification independent of the geographical region. ies also found an association with early disease onset, 32,33 which
Of the axSpA patients, 39.7% fulfilled the clinical arm of the was also suggested in our study, although it was not statistically
ASAS classification criteria. This finding is in agreement with other significant.
studies from Europe, such as the data from the DESIR French in- Although our study sample is relatively low, it is considered as
ception cohort, which showed that 39.8% of the patients fulfilled acceptable compared to the small Lebanese population (around 4
the clinical arm of the ASAS axSpA classification criteria; however, million people), and the 12 rheumatology practices (out of 45 ac-
this cohort included patients at an early stage of inflammatory back tive rheumatologists in the country) are fairly distributed across the
pain. 26 Lebanese territory and are representative of the target population.
The association with psoriasis could reflect a diagnostic bias as We deliberately decided not to include ethnic minorities such as pa-
patients with psoriasis and HLA‐B27 have a higher chance of being tients of Armenian descent, based on the assumption of different
diagnosed by the rheumatologist and of being identified in their genetic background and with the objective to keep the study popu-
axSpA database. However, it could also reflect a true clinical associ- lation as homogeneous as possible.
ation as was suggested in previous studies. 27,28 A small discordance was found between local and central labora-
The tendency of a negative association with IBD has been pre- tory testing (3 out of 88 tests, occurring in both directions). Although
viously suggested29-31 and deserves further investigation regard- this could be interpreted as a classification bias, the effect on the
ing the gene‐environment interaction hypothesis. Previous studies study results remains minimal.
ZIADE et al. | 713

In conclusion, our study confirmed a low prevalence of HLA‐B27 3. Poddubnyy D, Rudwaleit M. Early spondyloarthritis. Rheum Dis Clin
in axSpA and BD patients in the Lebanese population compared to North Am. 2012;38:387‐403.
4. Chaaya M, Slim ZN, Habib RR, et al. High burden of rheu-
northern populations. Despite the low prevalence, HLA‐B27 is char-
matic diseases in Lebanon: a COPCORD study. Int J Rheum Dis.
acterized by high specificity and a high positive LR, similar to those 2012;15(2):136‐143.
found in European studies, which makes this test highly valuable for 5. Amor B, Dougados M, Mijiyawa M. [Criteria of the classification of
axSpA diagnosis. spondylarthropathies]. Rev Rhum Mal Osteoartic. 1990;57(2):85‐89.
6. Dougados M, van der Linden S, Juhlin R, et al. The European
HLA‐B27 was associated with SpA family history and psoriasis.
Spondylarthropathy Study Group preliminary criteria for
There was a tendency toward a positive association with thoracic the classification of spondylarthropathy. Arthritis Rheum.
involvement and early symptom onset and a tendency toward a neg- 1991;34(10):1218‐1227.
ative association with IBD. 7. van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic
criteria for ankylosing spondylitis. A proposal for modification of
Of the axSpA patients, 39.7% fulfilled the clinical arm of the ASAS
the New York criteria. Arthritis Rheum. 1984;27(4):361‐368.
classification criteria. These findings are of utmost importance when 8. Baddoura R, Awada H, Okais J, et al. Validation of the European
planning local cost‐effective referral strategies, including HLA‐B27, Spondylarthropathy Study Group and B. Amor criteria for spondy-
diagnosing axSpA in a patient of Lebanese origin and planning local larthropathies in Lebanon. Rev Rhum Engl Ed. 1997;64(7–9):459–464.
9. Wendling D, Claudepierre P, Prati C. Early diagnosis and management
screening recommendations.
are crucial in spondyloarthritis. Jt Bone Spine. 2013;80(6):582‐585.
10. Rudwaleit M, van der Heijde D, Landewé R, et al. The development
of Assessment of SpondyloArthritis international Society classifica-
AC K N OW L E D G E M E N T S
tion criteria for axial spondyloarthritis (part II): validation and final
Blood Transfusion Center in Hotel‐Dieu de France Hospital, Beirut, selection. Ann Rheum Dis. 2009;68(part II):777‐783.
11. Rudwaleit M, Sieper J. Referral strategies for early diagnosis of axial
Lebanon. Laboratory team of the Medical University, Hannover,
spondyloarthritis. Nat Rev Rheumatol. 2012;8(5):262‐268.
Germany. Ms. Myriam Ziade, archeologist at the General Directorate 12. Sieper J, Rudwaleit M, Baraliakos X, et al. The Assessment of
of Antiquities in Lebanon. This study is funded by AbbVie SpondyloArthritis international Society (ASAS) handbook: a
Deutschland GmbH & Co. KG. AbbVie contributes to study logistics guide to assess spondyloarthritis. Ann Rheum Dis. 2009;68(Suppl
2):ii1‐i44.
and site management. The authors determined study design, con-
13. Reveille JD. HLA‐B27 and the Seronegative spondyloarthropathies.
duct of the study and the content of the publication. No payments Am J Med Sci. 1998;316(4):239‐249.
were made to the authors for writing this manuscript. 14. Khan MA. HLA‐B27 and its subtypes in world populations. Curr
Opin Rheumatol. 1995;7(4):263‐269.
15. Awada H, Baddoura R, Naman R, Klayme S. Weak association be-
C O N FL I C T O F I N T E R E S T tween HLA‐B27 and the spondylarthropathies in Lebanon. Arthritis
Rheum. 1997;40(2):388‐389.
None. 16. Ziade NR. HLA B27 antigen in Middle Eastern and Arab countries:
systematic review of the strength of association with axial spon-
dyloarthritis and methodological gaps. BMC Musculoskelet Disord.
AU T H O R C O N T R I B U T I O N 2017;18:280‐284.
17. Gofton JP, Bennett PH, Smythe HA, Decker JL. Sacroiliitis and
Nelly Ziade and Xenofon Baraliakos designed the study, supervised ankylosing spondylitis in North American Indians. Ann Rheum Dis.
the research, drafted and corrected the final manuscript. Torsten 1972;31(6):474‐481.
18. Khan M, Kushner I, Braun WE. Comparison of clinical features in
Witte and Pierre Ghorra did the laboratory testing and ensured re-
HLA‐B27 positive and negative patients with ankylosing spondyli-
sults’ quality. Iyad Mallak performed data collection and data analy-
tis. Arthritis Rheum. 1977;20(4):909‐912.
sis. Nelly Ziade, Ghada Abi Karam, Georges Meheb, Laure Irani, 19. Gran JT, Husby G, Hordvik M. Prevalence of ankylosing spondylitis
Elie Alam, Jamil Messaykeh, Jeanine Menassa, Kamel Mroue’, Imad in males and females in a young middle‐aged population of Tromsø,
Uthman and Abdel Fattah Masri recruited the patients for the study. northern Norway. Ann Rheum Dis. 1985;44(6):359‐367.
20. Haber M, Doumet‐Serhal C, Scheib C, et al. Continuity and ad-
All the authors contributed to the elaboration and review of the final
mixture in the last five millennia of Levantine history from ancient
manuscript. Canaanite and present‐day Lebanese Genome Sequences. Am J
Hum Genet. 2017;101(2):274‐282.
21. Collelo T.Lebanon: a country study. Washington: GPO for the
ORCID Library of Congress, editor;1987.
22. Statistics Lebanon. The Lebanese demographic reality (‫عقاولا‬
Nelly Ziade https://orcid.org/0000-0002-4479-7678 ‫ ) نانبل يف يفارغوميدلا‬25;2013.
23. Reece WS.Demographics of Lebanon. https://en.wikipedia.org/
wiki/Demographics_of_Lebanon. p. 2‐6.
REFERENCES 24. Zalloua PA, Platt DE, El Sibai M, et al. Identifying genetic traces of
historical expansions: Phoenician Footprints in the Mediterranean.
1. Reveille JD. Global Interfacing. Curr Opin Rheumatol. Am J Hum Genet. 2008;83(5):633‐642.
2013;25(4):468‐476. 25. Serre J‐L, Lefranc G, Loiselet J, Jacquard A. HLA mark-
2. Boonen A, Van Der LindenSM. The burden of ankylosing spondyli- ers in six Lebanese religious subpopulations. Tissue Antigens.
tis LIFE. J Rheumatol. 2006;33:4‐11. 1979;14(3):251‐255.
714 | ZIADE et al.

26. Bakker P, Moltó A, Etcheto A, et al. The performance of different 32. Awada H, Abi-Karam G, Baddoura R, Okais J, Attoui S. Clinical,
classification criteria sets for spondyloarthritis in the worldwide radiological, and laboratory findings in Lebanese spondylarthrop-
ASAS‐COMOSPA study. Arthritis Res Ther. 2017;19(1);96. athy patients according to HLA-B27 status. Joint Bone Spine.
27. Sheehan NJ. HLA‐B27: what’s new? Rheumatology. 2000;67(3):194–198.
2010;49(4):621‐631. 33. Mou Y, Zhang P, Li Q, et al. Clinical features in juvenile‐onset anky-
28. Ruiz DG, de Azevedo M, Lupi O. HLA‐B27 frequency in a losing spondylitis patients carrying different B27 subtypes. Biomed
group of patients with psoriatic arthritis. An Bras Dermatol. Res Int. 2015;2015:1‐5.
2012;87(6):847‐850.
29. Brakenhoff L, van der Heijde DM, Hommes DW, Huizinga T, Fidder
HH. The joint‐gut axis in inflammatory bowel diseases. J Crohn’s
How to cite this article: Ziade N, Abi Karam G, Merheb G, et al.
Colitis. 2010;4(3):257‐268.
HLA‐B27 prevalence in axial spondyloarthritis patients and in
30. Haas W, Meijers K, Tytgat G. Clinical characteristics and results of
histocompatibility typing ( HLA B27) in 50 patients with both anky- blood donors in a Lebanese population: Results from a
losing spondylitis and inflammatory bowel disease. Ann Rheum Dis. nationwide study. Int J Rheum Dis. 2019;22:708–714. https://
1978;37:36‐41. doi.org/10.1111/1756-185X.13487
31. Ajene AN, Fischer Walker CL, Black RE. Enteric pathogens and re-
active arthritis: a systematic review of Campylobacter, Salmonella
and Shigella‐associated reactive arthritis. J Heal Popul Nutr.
2013;31(3):299‐307.
Copyright of International Journal of Rheumatic Diseases is the property of Wiley-Blackwell
and its content may not be copied or emailed to multiple sites or posted to a listserv without
the copyright holder's express written permission. However, users may print, download, or
email articles for individual use.