Received: 17 January 2018 Revised: 9 February 2018 Accepted: 12 February 2018

DOI: 10.1002/rmv.1977

REVIEW

Clinical outcomes of current medical approaches for Middle

East respiratory syndrome: A systematic review and
meta‐analysis
Mostafa Ebraheem Morra1,2 | Le Van Thanh2,3 | Mohamed Gomaa Kamel2,4 |

Ahmed Abdelmotaleb Ghazy5 | Ahmed M.A. Altibi6 | Lu Minh Dat2,7 |

Tran Ngoc Xuan Thy2,3 | Nguyen Lam Vuong2,3 | Mostafa Reda Mostafa2,8 |
Sarah Ibrahim Ahmed9 | Sahar Samy Elabd10 | Samreen Fathima11 | Tran Le Huy Vu12 |

Ali S. Omrani13 | Ziad A. Memish14 | Kenji Hirayama15 | Nguyen Tien Huy16,17

1
Faculty of Medicine, Alazhar University, Cairo, Egypt
2
Online Research Club (http://www.onlineresearchclub.org/)
3
University of Medicine and Pharmacy at Ho Chi Minh City, Ho Chi Minh City, Vietnam
4
Faculty of Medicine, Minia University, Minya, Egypt
5
Department of Cardiology, Shebin El‐Kom Teaching Hospital, Shibin El Kom, Egypt
6
University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA
7
Pham Ngoc Thach University of Medicine, Ho Chi Minh City, Vietnam
8
Faculty of Medicine, Tanta University, Tanta, Egypt
9
Faculty of Medicine, Cairo University, Giza, Egypt
10
Faculty of Medicine, Benha University, Benha, Egypt
11
Deccan College of Medical Sciences, University of Health and Sciences, Hyderabad, India
12
University of California, Los Angeles, California, USA
13
Department of Medicine, Section of Infectious Diseases, King Faisal Specialist Hospital and Research Centre, Riyadh, Kingdom of Saudi Arabia
14
Ministry of Health, Riyadh, Saudi Arabia
15
Department of Immunogenetics, Institute of Tropical Medicine (NEKKEN), Leading Graduate School Program, and Graduate School of Biomedical Sciences, Nagasaki
University, Nagasaki, Japan
16
Evidence Based Medicine Research Group & Faculty of Applied Sciences, Ton Duc Thang University, Ho Chi Minh City, Vietnam
17
Department of Clinical Product Development, Institute of Tropical Medicine (NEKKEN), Leading Graduate School Program, and Graduate School of Biomedical
Sciences, Nagasaki University, Nagasaki, Japan

Correspondence
Nguyen Tien Huy, Evidence Based Medicine Summary
Research Group & Faculty of Applied Sciences, Middle East respiratory syndrome (MERS) is a respiratory disease caused by MERS
Ton Duc Thang University, Ho Chi Minh City,
Vietnam. coronavirus. Because of lack of vaccination, various studies investigated the therapeu-
Email: [email protected] tic efficacy of antiviral drugs and supportive remedies. A systematic literature search
Funding information
from 10 databases was conducted and screened for relevant articles. Studies reporting
Japan Initiative for Global Research Network
on Infectious Diseases (J‐GRID); Ministry of information about the treatment of MERS coronavirus infection were extracted and
Education, Culture, Sports, Science, and
analyzed. Despite receiving treatment with ribavirin plus IFN, the case fatality rate
Technology (MEXT) of Japan, Grant/Award
Number: 16H05844, 2016‐2019 was as high as 71% in the IFN‐treatment group and exactly the same in patients

List of abbreviations: CART, the classification and regression tree model; CHF, congestive heart failure; CI, confidential intervals; CRD, chronic renal disease; DM,
diabetes mellitus; ECMO, extracorporeal membrane oxygenation; MD, mean difference; MERS‐CoV, Middle East respiratory syndrome coronavirus; MERS, Middle
East respiratory syndrome; NPV, negative prediction value; PPV, positive prediction value; SARS, severe acute respiratory syndrome; SPSS, Statistical Package for
Social Sciences
Mostafa Ebraheem Morra, Le Van Thanh, and Mohamed Gomaa Kamel equally contributed in the work.

Rev Med Virol. 2018;e1977. wileyonlinelibrary.com/journal/rmv Copyright © 2018 John Wiley & Sons, Ltd. 1 of 9
https://doi.org/10.1002/rmv.1977
2 of 9 MORRA ET AL.

who received supportive treatment only. Having chronic renal disease, diabetes
mellitus and hypertension increased the risk of mortality (P < .05), and chronic renal
disease is the best parameter to predict the mortality. The mean of survival days from
onset of illness to death was 46.6 (95% CI, 30.5‐62.6) for the IFN group compared
with 18.8 (95% CI, 10.3‐27.4) for the supportive‐only group (P = .001). Delay in
starting treatment, older age group, and preexisting comorbidities are associated with
worse outcomes. In conclusion, there is no difference between IFN treatment and sup-
portive treatment for MERS patients in terms of mortality. However, ribavirin and IFN
combination might have efficacious effects with timely administration and monitoring
of adverse events. Large‐scale prospective randomized studies are required to assess
the role of antiviral drugs for the treatment of this high mortality infection.

KEY W ORDS

interferon, MERS‐CoV, Middle East respiratory syndrome, systematic review

1 | I N T RO D U CT I O N 2 | METHODS

Middle East respiratory syndrome (MERS) is a viral respiratory disease 2.1 | Search strategy and selection criteria
caused by MERS coronavirus (MERS‐CoV), and it is relatively new to
Our study was performed according to the recommendations of
humans.1 The first report took place in Saudi Arabia in 2012.2 After-
the Preferred Reporting Items for Systematic Reviews and Meta‐
wards, MERS cases were reported across a wide geographic distribu-
Analyses.13 We developed a protocol of methods and registered it in
tion around the globe. However, the vast majority of these cases
the international prospective register of systematic reviews (PROS-
were directly or indirectly linked to traveling or residing in the Middle
2 PERO) (reference, CRD42015024819). In June 2015, we conducted
East. There is no vaccine against MERS‐CoV, and current recommen-
a systematic search of 10 databases including PubMed, Scopus, Web
dations advise people to only adhere to general preventive measures
of Science, Google Scholar, WHO Global Health Library, Virtual Health
such as handwashing, frequently disinfecting touched surfaces, and
2 Library (containing Cochrane, MEDLINE, LILACS, IBECS, and SciELO),
avoiding personal contact with infected populations.
POPLINE, New York Academy of Medicine Grey Literature Report
The clinical picture of the infection ranges from asymptomatic or
(NYAMGLR), SIGLE, and ClinicalTrials.gov. Search results were limited
mild respiratory symptoms to severe acute respiratory disease and
to references published since January 1, 2012, when MERS was first
death.3 Most infected patients present with flu‐like symptoms (cough,
detected. In addition, the human filter was applied in the case of
shortness of breath, fever, and chills) then rapidly progress into severe
PubMed.
illness with pneumonia and acute respiratory distress symptoms, for
After removing duplicates, three trained reviewers were assigned
which many of them require mechanical ventilation (72%).4 Further-
to independently screen the titles and abstracts of all references gen-
more, the MERS can be complicated by acute renal failure that requires
erated from the aforementioned search strategy on the basis of the
hemodialysis and by disseminated intravascular coagulation with
2,4-9 following inclusion and exclusion criteria. Inclusion criteria were (a)
multiorgan failure contributing to the high mortality of the disease.
any study that gives information about the treatment of MERS‐CoV
Currently, there is no reliable remedy for MERS, and all interven-
infection, (b) all types of study designs were included, and (c) no restric-
tions range from supportive to antiviral therapy. Two in vitro studies
tion was made with respect to language, age, and area. Exclusion
suggested a possible efficacious effect of interferon alpha‐2b and riba-
10,11 criteria were (a) data that could not be reliably extracted, (b) data sets
virin in the treatment of MERS infection. Consequently, the inves-
considered as overlapping, (c) studies published before 1/1/2012, (d)
tigators further examined the efficacy of these drugs in an animal
book chapter, thesis, letter, conference paper, poster, or editorial,
study.12 Potential benefits of these antiviral treatments in both
and (e) animal or in vitro studies. Three reviewers compared their
in vitro and animal studies persuaded clinicians to question the feasibil-
screening results and discussed the differences. A consensus was
ity and applicability of such an approach in humans. Therefore, we
reached through discussion.
aimed to recapitulate the evidence from all human published data
about MERS clinical management in a systematic review and meta‐
analysis, to summarize the efficacy and safety of current applied ther-
apeutics and define risk factors associated with outcomes. As a result,
2.2 | Data extraction
we may provide a better approach to more compatible management of Extracted data included publication year, year of research, country and
fatal consequences of MERS infections and the risk imposed by recent city of the patients, year of subject recruitment, study design, partici-
outbreaks in densely populated areas. pant enrollment, data collection method, baseline characteristics
MORRA ET AL. 3 of 9

before treatment, diagnostic method of MERS‐CoV, time from admis-

sion to treatment start, treatment for MERS‐CoV, and outcome sur-
vival. This work was conducted by 2 investigators evaluating the
references independently, and all disagreements were discussed to
reach a consensus from supervisors.

2.3 | Quality assessment

The quality of included clinical data was assessed using (CARE) state-
ment for case reports14 and the 9 metrics tool for nonrandomized
studies.15 Three reviewers were assigned to assess each included ref-
erence independently.

2.4 | Statistical analysis

Mortality rates were treated as dichotomous variables with their
respective 95% confidential intervals (CI). Statistical heterogeneity
FIGURE 1 Preferred Reporting Items for Systematic Reviews and
was assessed using the I2 statistic16,17 and assumed to be influential Meta‐Analyses flow diagram of studies' screening and selection
when I2 was greater than 50% or P ≤ .1.18,19 A fixed‐effect model according to inclusion and exclusion criteria
was used because there was no evidence of heterogeneity between
5 references were identified through manual search.7-9,32,33 So a total
studies. Meta‐analysis was performed using data analysis and statisti-
of 16 studies were eligible for selection as shown in Figure 1.
cal software (STATA) that was developed by StataCorp. Fisher exact
(or chi‐square, as appropriate) and Mann‐Whitney U tests were used
for the categorical and continuous variables, respectively. The classifi- 3.2 | Characteristics of included studies
cation and regression tree (CART) model was used to identify indepen-
Of the 16 studies including 116 patients, 10 were case
dent variables that predict mortality outcome.20 Invasive ventilation
reports,7-9,21,24,25,28,30,31,33 2 were case series,23,26 and 4 were obser-
and renal replacement therapy were also chosen as the outcomes in
vational studies.22,27,29,32 Saudi Arabia was the country of origin for
the CART model because of direct correlation with severity and mor-
patients in 13 studies, and the rest were from France, Greece, and
tality.20 All possible variables were extracted to build CART
Qatar. Detailed characteristics of patients in our included articles are
(Table S4). The performance parameters are accuracy, sensitivity, spec-
presented in Table 1.
ificity, positive predictive value (PPV), and negative predictive value
Eight studies used specific antiviral treatment21,22,25,26,28-30,32
(NPV).20 Results were considered to have statistical significance if
while 7 studies used supportive treatment (including invasive ventila-
the P value was <.05. For evaluating the survival rate over time
tion, prone position, renal replacement therapy, vasopressors, cortico-
between 2 groups of all‐type IFN and supportive‐only treatment, log‐
steroids, immunoglobulins, and oseltamivir).7-9,23,24,31,33 Omrani et al27
rank test and Kaplan‐Meier survival analysis were performed. We per-
used both specific antiviral treatment and supportive treatment. The
formed 2 sets of Kaplan‐Meier survival curves. The first set described
specific antiviral treatments were IFN (alpha‐2a, alpha‐2b, and beta‐
the survival time from hospital admission to death while the second set
1a), ribavirin, and several others including tenofovir, emtricitabine,
reported the survival time from onset of symptoms to death. The mean
lopinavir, and ritonavir. Among 116 patients recruited, 29 were
survival days was calculated for both sets. Patients who were still on
reported with detailed information regarding baseline characteristics,
admission while the original study ended were included in Kaplan‐
comorbidities, treatments, outcomes, and some with survival time,
Meier curves but considered as expired patients from final analysis.
which was subsequently used to perform univariable analysis and
Data were analyzed using statistical computing and graphics (R) soft-
Kaplan‐Meier survival curves (Table S4).
ware version 3.3.2 developed by R Foundation for Statistical Comput-
Hemodialysis dependency appeared in higher frequency in the IFN
ing. Statistical Package for Social Sciences (SPSS) released by
group than in the supportive‐only group (P = .025). Conversely, renal
International Business Machines Corporation was used for analysis as
replacement therapy and vasopressors were used more often in the
well.
supportive‐only group than in the IFN group (P = .019) (Table 2).

3 | RESULTS 3.3 | Quality assessment

The quality of 10 case reports7-9,21,24,25,28,30,31,33 and 2 case series23,26
3.1 | Search results was assessed using the CARE checklist (Table S1). All of them
A total of 1095 references were retrieved. Upon screening them contained an introduction and described the importance of the case
regarding inclusion and exclusion criteria from Section 2, eleven refer- in the abstract, main introduction section, demographics and symp-
ences were included for data extraction and analysis.21-31 Additionally, toms of the patients, significant clinical findings, timeline for patients'
4 of 9 MORRA ET AL.

TABLE 1 Characteristics of the included studies

Total Number of
Study Name Study Design Country Age Male Types of Treatment Cases Deaths

AlGhamdi et al21 Case report Saudi Arabia 44 1 Interferon 1 1

Al‐Tawfiq et al22 Retrospective Saudi Arabia 57.6 3 Interferon 5 5
observational study
Shalhoub et al29 Retrospective cohort Saudi Arabia 65.9 14 Interferon 24 18
Khalid et al26 Case series Saudi Arabia 58.8 5 Interferon 6 3
Khalid et al25 Case report Saudi Arabia 47 1 Interferon 2 0
Spanakis et al30 Case report Greece 69 1 Interferon 1 1
27
Omrani et al Retrospective cohort Saudi Arabia 65.5 32 Interferon (cases) and supportive 44 34
treatment only (control)
Shalhoub et al28 Case report Saudi Arabia 51 1 Interferon 1 0
Al‐Hameed et al32 Prospective cohort Saudi Arabia 56.5 6 Interferon 8 6
23
Arabi et al Case series Saudi Arabia 59 8 Supportive treatment only 12 7
Thabet et al31 Case report Saudi Arabia 9‐month‐old 1 Supportive treatment only 1 1
Guberina et al24 Case report Qatar 45 1 Supportive treatment only 1 0
Omrani et al33 Case report Saudi Arabia 43.3 3 Supportive treatment only 3 2
7
Guery et al Case report France 57.5 2 Supportive treatment only 2 1
Memish et al8 Case report Saudi Arabia 39 4 Supportive treatment only 4 2
Zaki et al9 Case report Saudi Arabia 60 1 Supportive treatment only 1 1

TABLE 2 Comparison between IFN and comparator groups (no The quality of 4 observational studies22,27,29,32 was assessed using
antiviral/IFN) the 9 metrics tool15 (Table S2). Three of them had a score23,27,29 of 5 of
Treatment with Comparator 9 as they described study design, characteristics of the patient popula-
Factor IFN (n = 68) Group (n = 48) P Value tion, inclusion criteria, method quality, and MERS diagnosis. However,
Age (mean) 62.2 57.3 none of the 3 studies described data collection method, assignment
Gender (males) 48 (71) 40 (83) .114 method of patients, exclusion criteria, and interpretation.
Mortality 48 (71) 34 (71) 1
Diabetes 38 (56) 26 (54) .862
Hypertension 26 (38) 9 (19) .024
3.4 | Treatment with antiviral drugs
Chronic renal impairment 22 (32) 11 (23) 0.27 Treatment with antiviral drugs, other than oseltamivir, was reported in
a
Dialysis dependent 10 (15) 1 (2) .025 9 studies. IFNs (IFN alpha‐2a, alpha‐2b, or beta‐1a) in combination
Congestive heart failure 17 (25) 10 (21) .603 with ribavirin were the common remedies used in all 9 studies.
Other comorbidities 23 (34) 3 (6) .0005a IFN alpha‐2a was used in 4 studies (n = 35 patients),21,27-29 all of
Number of comorbidities 136 60 which used a dose of 180 μg/wk for treatment. IFN alpha‐2b was used
Invasive ventilation 52 (76) 43 (90) .071 in 5 studies (n = 22)22,25,26,30,32 with a dose ranging between 100 and
Renal replacement 17 (25) 22 (46) .019 180 μg/wk. IFN alpha‐2b dose was not reported in Al‐Hameed et al.32
Corticosteroids 28 (41) 28 (58) .068 IFN beta‐1a was used in 2 studies (n = 12), both of which used a dose
Oseltamivir 31 (46) 31 (65) .043 of 44 μg/wk for treatment.28,29
Vasopressors 25 (37) 32 (67) .002 Ribavirin administration was started with a loading dose followed
Prone position 4 (6) 6 (12.5) .315a by subsequent doses in all 9 studies. The loading dose was 2000 mg
ECMO 3 (3) 5 (10.4) .272a for all studies while 400 mg in one study of AlGhamdi et al.21 The sub-

Abbreviation: ECMO, extracorporeal membrane oxygenation; IFN,

sequent doses, however, were variable among studies and ranged
interferon. between 400 and 3600 mg/d. The frequency of administration of riba-
Significant values are in bold. virin was 3 doses per day in 2 studies22,30 and 2 doses per day in
a
Fisher exact test. another 3 studies.21,28,29 Duration of treatment with ribavirin was also
variable and ranged between 5 and 26 days. The subsequent oral
ribavirin dose was adjusted according to the calculated creatinine
history organization, diagnostic methods, type of treatment, clinical clearance in 3 studies.25-27 However, the duration of treatment
outcomes, discussion of the relevant medical literature, rationale for (8‐10 days) and the loading dose (2000 mg) used in the 3 studies
conclusion, and the primary take away lessons from these case reports. were similar, regardless of creatinine clearance (Table S3).
However, none of them described the diagnostic challenges, prognos- In addition to treatment with IFN and ribavirin, the treatment reg-
tic characteristics, or patients' perspective. Only Guery et al7 reported imen in Shalhoub et al28 included treatment with tenofovir/
obtaining informed consent from patients. emtricitabine (TDF/FTC) 300/200 mg orally once daily, in combination
MORRA ET AL. 5 of 9

with ritonavir‐boosted atazanavir (atazanavir 300 mg plus ritonavir 116, 81.9%) required the use of invasive mechanical ventilation:
100 mg) orally once daily. Similarly, triple therapy with IFN alpha‐2a, 76.4% of patients in IFNs treatment group (52 of 68) compared with
ribavirin, and lopinavir/ritonavir (400/100 mg twice daily) was used 90% of patients in supportive treatment–only group (43 of 48).
to treat the Greek patient in Spanakis et al.30
It is worth mentioning that there was a significant lag period 3.6 | Therapeutic outcomes
between presentation and initiation of the antiviral treatment. The Regarding mortality, studies with more than 2 cases were included in
period from the admission of patients to the initiation of antiviral (IFNs the meta‐analysis. The pooled proportion of mortality was 0.714
and ribavirin) was reported in 13 patients and had a mean of 12 days (0.618‐0.795) from 8 studies including 106 MERS patients (Figure 2).
and ranged between 1 day in Khalid et al26 and 21 days in Al‐Tawfiq In 68 patients received IFN treatment, the mortality rate was high
et al (Table S3).22 (71%) in spite of receiving treatment with ribavirin plus IFN (alpha‐
2a, alpha‐2b, or beta‐1a). Likewise, the mortality rate was high in
3.5 | Supportive treatment patients who received supportive treatment only (71%, n = 48). There
Supportive treatment alone was used in 7 studies (n = 24 patients), was no statistical significant mortality difference when comparing mor-
whereas supportive treatment along with antiviral medications was tality of both groups (P = 1) (Table 2). The same insignificant mortality
used in another 8 studies (n = 48) and one study compared support- difference was shown when comparing 3 types of IFN treatments: IFN
ive/antiviral medications against supportive treatment alone (n = 20 alpha‐2a, IFN alpha‐2b, and IFN beta‐1a (P = .65) (Table 3).
and 24, respectively). Different modalities of treatments were used The duration from hospitalization to death was reported in 78 of
to support the patients, including invasive ventilation, extracorporeal 82 deceased patients. All 78 reported patients died within 3 months
membrane oxygenation, corticosteroids, renal replacement therapy, of hospitalization. Moreover, 59% of the 82 patients died within the
vasopressors, oseltamivir, and prone positioning. Most patients (95 of first 2 weeks of admission, and 79% died within 1 month of admission.

FIGURE 2 Forest plot meta‐analysis of 8 studies regarding the mortality rate of patients with MERS. CI, Confidence interval; P, probability value

TABLE 3 Comparison between types of IFNs

Factor IFN Alpha‐2a (n = 34) IFN Alpha‐2b (n = 22) IFN Beta‐1a (n = 11) P Value
Age (mean) 65.8 57.1 67
Gender (males) 27 (79) 16 (73) 4 (36) .024
Mortality 26 (76) 15 (68) 7 (64) .65
Diabetes 24 (71) 9 (41) 5 (45) .065
Hypertension 11 (32) 8 (36) 7 (64) .173
Chronic renal impairment 11 (32) 8 (36) 3 (27) .869
Dialysis dependent 4 (12) 4 (18) 2 (18) .723a
Congestive heart failure 10 (29) 5 (23) 2 (18) .715
Other comorbidities 13 (3) 10 (9) 0 .0027
Invasive ventilation 30 (85) 16 (73) 6 (55) .052
Renal replacement therapy 12 (35) 5 (23) … .317
Corticosteroids 11 (32) 16 (73) … .003
Oseltamivir 18 (53) 13 (59) … .655
Vasopressors 14 (41) 11 (50) … .517
Prone position 4 (12) 0 … .146a
ECMO 2 (6) 1 (5) … 1 a

Abbreviation: ECMO, extracorporeal membrane oxygenation; IFN, interferon.

Significant values are in bold.
a
Fisher exact test.
6 of 9 MORRA ET AL.

In a case series of 6 patients from Saudi Arabia by Khalid et al,26 3 only 2 patients of 12 survived patients (P < .001). For continuous var-
patients had comorbid conditions and needed mechanical ventilation iables, there was a significant difference between death and survival in
before dying, while the other 3 patients who did not require mechan- age, where the dead patients were older, and time from onset of illness
ical ventilation survived. A case report from Greece described one to initiation of antiviral treatment, being longer in the deceased
patient with MERS‐CoV infection who died despite receiving IFN patients (P < .05) (Table 4).
alpha‐2a, ribavirin, and lopinavir. The patient had a multiorgan failure
and was later diagnosed with colon cancer. MERS‐CoV was not detect- 3.8 | CART model
able in his respiratory tract until several days before death.30
The modeling tool, CART, identified CRD as the best parameter to pre-
Regarding the adverse effects of interferon and ribavirin combina-
dict mortality (Figure 3A). The performance of the decision tree tool
tion therapy, 2 patients had pancreatic enzyme elevation, and one had
that classified mortality outcome was at an accuracy of 72.4%, sensi-
significant hemolysis when being treated with IFN alpha‐2a plus ribavi-
tivity of 52.9%, specificity of 100%, PPV of 100%, and NPV of 60%.
rin.22 In addition, the reduction in hemoglobin level was 4.32 g/L in 20
In addition, treatment with inotropes was exhibited as the best param-
MERS patients treated with IFN alpha‐2b plus ribavirin, whereas
eter to predict renal replacement therapy outcome (Figure 3B). The
Omrani et al27 reported only 2.14 g/L of hemoglobin reduction in 24
performance of the decision tree tool that classified renal replacement
patients treated with supportive‐only treatment (P = .002).
therapy outcome was at an accuracy of 86.2%, sensitivity of 66.67%,
specificity of 95%, PPV of 85.7%, and NPV of 86.4%. No significant
3.7 | Risk factors associated with outcomes
results were detected regarding the invasive ventilation outcome.
There was a significant difference between death and survival in
patients with chronic renal disease (CRD). Nine CRD patients died of
17 deceased cases, and no patient had CRD in the survival group. In
3.9 | Survival rate over time
addition, diabetes mellitus (DM) and hypertension showed similar sig- Survival time from admission to death was compared for IFN‐treated
nificant variation, suggesting that having CRD, hypertension, and/or (n = 30) and for non‐IFN (supportive care only, n = 14) patients. All
DM increased the risk of mortality (P < .05). There was no significant 44 cases died within 80 days after hospital admission (Figure 4A). Only
difference between death and survival regarding gender, ribavirin, cor- one female case treated with IFN was eliminated from analysis
ticosteroid, oseltamivir, IFN beta‐1a, IFN alpha‐2b, IFN alpha‐2a, con- because the patient remained intubated when the original study
gestive heart failure, and other comorbidities (P > .05). All 17 deceased ended. However, she had met death criteria; thus, she was included
patients required mechanical ventilation before dying compared with with deceased patients in the final outcome analysis. The mean

TABLE 4 Associated factors with cases' outcomes

Variables Deaths (n = 17)a Survivors (n = 12)a MD (95% CI) OR (95% CI) P Value

Age, mean (SD) 56.4 (21.6) years 39.8 (13.3) years 16.6 (3.3‐30.0) … .014
Time from admission to antiviral 15.1 (4.4) days 1.7 (0.6) days 13.4 (10.2‐16.6) … .003
treatment start, mean (SD)
Gender (male), % 15 (88) 10 (83) 1.5 (0.18‐12.46) 1
DM 7 0 … .023
HTN 6 0 … .028
CRD 9 0 … .003
DD 4 0 … .12
CHF 2 0 … .498
Other comorbidities 10 3 4.29 (0.84‐21.76) .13
IFN alpha‐2a 1 1 0.69 (0.04‐12.20) 1
IFN alpha‐2b 9 5 1.58 (0.35‐7.00) .71
IFN beta‐1a 1 0 … .414
Ribavirin 10 6 1.43 (0.32‐6.32) .716
Ventilation 17 2 … <.001
Corticosteroid 8 6 0.89 (0.20‐3.90) 1
Oseltamavir 10 4 2.86 (0.61‐13.34) .264
Inotropes 6 1 6.00 (0.62‐58.43) .187
Renal therapy 8 1 9.80 (1.02‐93.50) .043

Abbreviations: CHF, congestive heart failure; CI, confidential interval; CRD, chronic renal diseases; DD, dialysis dependent; DM, diabetes mellitus; HTN,
hypertension; MD, mean difference; SD, standard deviation.
Other comorbidities: colon adenocarcinoma, HIV, renal transplant, asthma, sleep apnea, coronary artery disease, atrial fibrilation, ischemic heart disease,
right bundle branch block, cardiomyopathy, MI, dyslipidemia, histamine induced angioedema, multiple myeloma, and obesity.
Significant values are in bold.
a
Numbers are the frequency and percent (%) otherwise stated.
MORRA ET AL. 7 of 9

FIGURE 3 The classification and regression tree model of 29 cases with individual data. A, Chronic renal disease was the best prediction variable
for mortality rate. B, Inotropes was best prediction variable for renal replacement therapy

FIGURE 4 Kaplan‐Meier survival curves showing death days over time in IFN‐treated patients and supportive‐only groups. A, Death days from
hospital admission to death for 44 cases (P = .977) and, B, death days from onset of illness to death for 13 cases (P = .001)

survival days was 21.3 days (95% CI, 14.1‐28.5) for IFN group and Further specific comparison of death versus nondeath cases in
21.4 days (95% CI, 12.4‐30.4) for the supportive‐only group, both IFN and supportive treatments was conducted to get the mean
P = .977. Overall, mean of survival days for both groups was 21.3 days from onset to admission. The mean of days for 10 IFN‐treated
(95% CI, 15.7‐26.9). dead patients and 5 IFN‐treated survivals was 7.79 and 4.40, respec-
Only 7 cases in the IFN group and 6 cases in the supportive‐only tively, with MD (95% CI) = 3.39 (−3.48 to 10.26), P = .3. Moreover,
group reported information about time from onset of symptoms to the mean of days for 13 supportively treated dead patients and 10
death. A second Kaplan‐Meier survival analysis was performed and supportively treated survivals was 2.69 and 2.1, respectively, with
showed that all of them died within 89 days from onset of illness. The MD (95% CI) = .59 (−1.35 to 2.53), P = .53. The mean of days from
mean of survival days was 46.6 (95% CI, 30.5‐62.6) for the IFN group onset of illness to admission for pooled total cases of death and sur-
compared with 18.8 (95% CI, 10.3‐27.4) for the supportive‐only group. vived was 4.07 and 2.27, respectively, with MD (95% CI) = 1.8
The difference between the 2 groups was statistically significant (−0.81 to 4.41), P = .17.
(P = .001) (Figure 4B). The longer survival period from onset to death
was simply attributed to the duration between onset and admission.
The mean time from onset of illness to the admission of 23 4 | DISCUSSION
patients in IFN group was 6.52 days compared with the same number
of patients in the supportive‐only group who had a mean of 2.43 days. Our systematic review highlights the significance of age and period
The mean difference (MD) of time between 2 groups was also calcu- between the illness onset and start of antiviral therapy in MERS cases'
lated: MD (95% CI) = 4.09 (2.71‐5.47), P < .05. prognostic assessment. Our results revealed that younger age and
8 of 9 MORRA ET AL.

more rapid initiation of antiviral therapy are associated with higher 5 | CO NC LUSIO N
chances of survival. We also found that patients who require ventila-
tion are more likely to die compared with patients who do not require There is no evidence of any difference between IFN treatment and
it. Furthermore, there is no difference between IFN treatment and sup- supportive treatment for MERS patients in terms of mortality. The
portive treatment in terms of mortality rate and no significant effect on ribavirin and IFN combination might have promising effects where
the duration of survival in infected patients' different groups. therapy can be started promptly and adverse effects monitored care-
Several factors can affect the clinical outcomes of MERS therapy. fully; a randomized controlled trial is required to assess this possibility.
In our data, there was a significantly higher number of patients with Concerning prognostic factors, delayed treatment, older age, and
hypertension and dialysis dependence in the IFN group compared with accompanying comorbidities such as hypertension, DM, chronic kidney
the supportive‐only group. The survival time of patients receiving disease, and dialysis dependence are associated with worse outcomes.
antiviral therapy and the efficacy of the drugs were likely reduced Because of high fatality, the seriousness of this newly emerging dis-
because of hypertension and dialysis dependence. Nevertheless, upon ease, and a limited number of available cases, we believe there is an
investigating the outcomes of 13 patients under treatment for MERS, urgent need for large‐scale clinical trials on the efficacy of antiviral
we found that, despite the higher number of comorbidities in the IFN treatment of MERS‐CoV infections.
group compared with the supportive‐only group, patients on IFN
survived considerably longer from onset of symptoms than those on CONFLIC T OF IN TE RE ST
supportive therapy. This result highlights the potential efficacy of The authors declare no competing interests.
antiviral treatment in MERS‐CoV infection. However, the number of
cases is small to draw any conclusions, and further studies are FUNDING
definitely required.
This work was supported in part by a “Grant‐in‐Aid for Scientific
In terms of different supportive remedies, corticosteroids did not
Research (B)” (16H05844, 2016‐2019, for Nguyen Tien Huy) from
show any promising effects on the survival of infected patients, which
the Ministry of Education, Culture, Sports, Science, and Technology
is also consistent with previous studies.8,23,27,33 Patients with signs of
(MEXT) of Japan and by the Japan Initiative for Global Research Net-
severe respiratory distress, shock, or hypoxemia should be given oxy-
work on Infectious Diseases (J‐GRID) for Kenji Hirayama. The funders
gen therapy immediately, and those who cannot maintain a SpO2 ≥ 90%
had no role in the study design, data collection and analysis, decision to
with oxygen therapy should be considered for intubation and mechan-
publish, or preparation of the manuscript.
ical ventilation. However, noninvasive ventilation should be avoided
because of the high risk of spreading the infection.34 Patients who
AUTHOR CONTRIBUTIONS
have an indication for ventilation were more likely to have an accom-
panying serious condition compared with other patients,34 which could N.T.H., A.A.G., M.E.M., and K.H. participated in the design of the study.

explain the higher risk of mortality among those patients as shown in L.V.T., L.M.D., M.G.K., N.L.V., and A.M.A.A. performed in the analysis

our results. and interpreted it. All authors contribute the screening and data

Because of high case fatality rate and faster progression to respi- extraction, wrote the manuscript, read, and approved the final

ratory failure, it will be pressing to distinguish MERS from severe acute manuscript.

respiratory syndrome for timely management. However, it is consid-

ered a diagnostic challenge for health care professionals due to similar ORCID

clinical features.35 Our results obtained with CART modeling suggest Ahmed Abdelmotaleb Ghazy http://orcid.org/0000-0002-9145-
that chronic respiratory disease can be the best predictor of death in 7115
patients with MERS. Therefore, it is of utmost necessity to thoroughly Nguyen Tien Huy http://orcid.org/0000-0002-9543-9440
assess and continuously monitor the respiratory functions of patients
with suspected MERS‐CoV infection. Also, health care agencies in high RE FE RE NC ES
risk areas are recommended to bridge any gaps in their medical facili- 1. Kupferschmidt K. Infectious diseases. MERS surges again, but pan-
demic jitters ease. Science. 2015;347(6228):1296‐1297. https://doi.
ties and personnel. They should get their staff familiar with the
org/10.1126/science.347.6228.1296
precautious measures that are required to face any future outbreaks.
2. WHO. Middle East respiratory syndrome coronavirus (MERS‐CoV).
Our study is limited by the small number of included articles and
May 2017.
patients, which hindered our ability to validate our CART model. In
3. Memish ZA, Al‐Tawfiq JA, Makhdoom HQ, et al. Screening for Middle
addition, there are no published clinical trials, and even only one ongo- East respiratory syndrome coronavirus infection in hospital patients
ing clinical trial has been found, testing the combination of lopinavir/ and their healthcare worker and family contacts: a prospective descrip-
ritonavir and IFN beta‐1b therapy on MERS patients tive study. Clin Microbiol Infect. 2014;20(5):469‐474.

(NCT02845843). Another recent trial has been registered but not yet 4. Assiri A, Al‐Tawfiq JA, Al‐Rabeeah AA, et al. Epidemiological, demo-
started recruitment, to investigate the safety and immunogenicity of graphic, and clinical characteristics of 47 cases of Middle East
respiratory syndrome coronavirus disease from Saudi Arabia: a descrip-
MERS‐CoV vaccine (NCT03399578). Nevertheless, further investiga- tive study. Lancet Infect Dis. 2013;13(9):752‐761.
tions are required to objectively assess the effect of antiviral therapy
5. Assiri A, McGeer A, Perl TM, et al. Hospital outbreak of Middle East
on the survival duration and overall outcomes of MERS‐CoV respiratory syndrome coronavirus. New England Journal of Medicine.
infections. 2013;369(5):407‐416.
MORRA ET AL. 9 of 9

6. Drosten C, Seilmaier M, Corman VM, et al. Clinical features and virolog- 24. Guberina H, Witzke O, Timm J, et al. A patient with severe respiratory fail-
ical analysis of a case of Middle East respiratory syndrome coronavirus ure caused by novel human coronavirus. Infection. 2014;42(1):203‐206.
infection. Lancet Infect Dis. 2013;13(9):745‐751. 25. Khalid M, Al Rabiah F, Khan B, Al Mobeireek A, Butt TS, Al Mutairy E.
7. Guery B, Poissy J, el Mansouf L, et al. Clinical features and viral diagno- Ribavirin and interferon‐alpha2b as primary and preventive treatment
sis of two cases of infection with Middle East respiratory syndrome for Middle East respiratory syndrome coronavirus: a preliminary report
coronavirus: a report of nosocomial transmission. The Lancet. of two cases. Antivir Ther. 2015;20(1):87‐91.
2013;381(9885):2265‐2272. 26. Khalid M, Khan B, Al Rabiah F, et al. Middle Eastern respiratory syn-
8. Memish ZA, Zumla AI, Al‐Hakeem RF, Al‐Rabeeah AA, Stephens GM. drome corona virus (MERS CoV): case reports from a tertiary care
Family cluster of Middle East respiratory syndrome coronavirus infec- hospital in Saudi Arabia. Ann Saudi Med. 2014;34(5):396‐400.
tions. N Engl J Med. 2013;368(26):2487‐2494. 27. Omrani AS, Saad MM, Baig K, et al. Ribavirin and interferon alfa‐2a for
9. Zaki AM, van Boheemen S, Bestebroer TM, Osterhaus AD, Fouchier severe Middle East respiratory syndrome coronavirus infection: a retro-
RA. Isolation of a novel coronavirus from a man with pneumonia in spective cohort study. Lancet Infect Dis. 2014;14(11):1090‐1095.
Saudi Arabia. N Engl J Med. 2012;367(19):1814‐1820. 28. Shalhoub S, AlZahrani A, Simhairi R, Mushtaq A. Successful recovery of
10. de Wilde AH, Raj VS, Oudshoorn D, et al. MERS‐coronavirus replica- MERS CoV pneumonia in a patient with acquired immunodeficiency
tion induces severe in vitro cytopathology and is strongly inhibited syndrome: a case report. J Clin Virol. 2015;62:69‐71.
by cyclosporin A or interferon‐α treatment. J Gen Virol. 2013; 29. Shalhoub S, Farahat F, Al‐Jiffri A, et al. IFN‐alpha2a or IFN‐beta1a in
94(8):1749‐1760. combination with ribavirin to treat Middle East respiratory syndrome
11. Falzarano D, de Wit E, Martellaro C, Callison J, Munster VJ, Feldmann coronavirus pneumonia: a retrospective study. J Antimicrob Chemother.
H. Inhibition of novel β coronavirus replication by a combination of 2015;70(7):2129‐2132.
interferon‐α2b and ribavirin. Sci Rep. 2013;3:1686. 30. Spanakis N, Tsiodras S, Haagmans BL, et al. Virological and serological
12. Falzarano D, De Wit E, Rasmussen AL, et al. Treatment with interferon‐ analysis of a recent Middle East respiratory syndrome coronavirus
[alpha] 2b and ribavirin improves outcome in MERS‐CoV‐infected infection case on a triple combination antiviral regimen. Int J Antimicrob
rhesus macaques. Nat Med. 2013;19(10):1313‐1317. Agents. 2014;44(6):528‐532.
13. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for 31. Thabet F, Chehab M, Bafaqih H, Al Mohaimeed S. Middle East respiratory
reporting systematic reviews and meta‐analyses of studies that evalu- syndrome coronavirus in children. Saudi Med J. 2015;36(4):484‐486.
ate health care interventions: explanation and elaboration. PLoS Med. 32. Al‐Hameed F, Wahla AS, Siddiqui S, et al. Characteristics and outcomes
2009;6(7):e1000100. of Middle East respiratory syndrome coronavirus patients admitted to
14. Case report guidelines [http://www.care‐statement.org/] an intensive care unit in Jeddah, Saudi Arabia. J Intensive Care Med.
15. The Newcastle‐Ottawa Scale (NOS) for assessing the quality of non- 2016;31(5):344‐348.
randomized studies in meta‐analyses [http://www.ohri.ca/programs/ 33. Omrani AS, Matin MA, Haddad Q, Al‐Nakhli D, Memish ZA, Albarrak
clinical_epidemiology/oxford.asp] AM. A family cluster of Middle East respiratory syndrome coronavirus
16. DerSimonian R, Laird N. Meta‐analysis in clinical trials. Control Clin infections related to a likely unrecognized asymptomatic or mild case.
Trials. 1986;7(3):177‐188. Int J Infect Dis. 2013;17(9):e668‐e672.

17. Higgins J, Thompson SG, Deeks JJ, Altman DG. Measuring inconsis- 34. CDC. Infection prevention/control and management guidelines for
tency in meta‐analyses [journal article as teaching resource, deposited patients with Middle East respiratory syndrome coronavirus (MERS
by John Flynn]. Br Med J. 2003;327:557‐560. CoV) infection. 2014.

18. Munafò MR, Flint J. Meta‐analysis of genetic association studies. 35. Hui DS, Memish ZA, Zumla A. Severe acute respiratory syndrome vs.
Trends Genet. 2004;20(9):439‐444. the Middle East respiratory syndrome. Curr Opin Pulm Med. 2014;
20(3):233‐241.
19. Zintzaras E, Lau J. Synthesis of genetic association studies for pertinent
gene‐disease associations requires appropriate methodological and sta-
tistical approaches. J Clin Epidemiol. 2008;61(7):634‐645.
SUPPOR TI NG INF ORMATI ON
20. Kamel MG, Nam NT, Han NHB, et al. Post‐dengue acute disseminated
encephalomyelitis: a case report and meta‐analysis. PLoS Negl Trop Dis. Additional Supporting Information may be found online in the
2017;11(6):e0005715. supporting information tab for this article.
21. AlGhamdi M, Mushtaq F, Awn N, Shalhoub S. MERS CoV infection in
two renal transplant recipients: case report. Am J Transplant.
2015;15(4):1101‐1104.
22. Al‐Tawfiq JA, Momattin H, Dib J, Memish ZA. Ribavirin and inter- How to cite this article: Morra ME, Van Thanh L, Kamel MG,
feron therapy in patients infected with the Middle East respiratory et al. Clinical outcomes of current medical approaches for
syndrome coronavirus: an observational study. Int J Infect Dis. 2014;
Middle East respiratory syndrome: A systematic review and
20:42‐46.
meta‐analysis. Rev Med Virol. 2018;e1977. https://doi.org/
23. Arabi YM, Arifi AA, Balkhy HH, et al. Clinical course and outcomes of
critically ill patients with Middle East respiratory syndrome coronavirus 10.1002/rmv.1977
infection. Ann Intern Med. 2014;160(6):389‐397.