Journal of the Neurological Sciences 368 (2016) 292–299

Contents lists available at ScienceDirect

Journal of the Neurological Sciences

journal homepage: www.elsevier.com/locate/jns

Review article
Case fatality of adult tetanus in Africa: Systematic review
and meta-analysis
Yohannes W. Woldeamanuel a,b,c,d,⁎, Adel T. Andemeskel e, Kwame Kyei d,
Meheret W. Woldeamanuel a, Woubishet Woldeamanuel a
a
Advanced Clinical Consultation & Research Center, Addis Abäba, Ethiopia
b
Department of Neurology and Neurological Sciences, Stanford University School of Medicine, Stanford, USA
c
Department of Neurology, Addis Abäba University School of Medicine, Addis Abäba, Ethiopia
d
Propria Health Solutions Co., USA
e
Department of Biology, Stanford University, Stanford, USA

a r t i c l e i n f o a b s t r a c t

Article history: Introduction: Tetanus is a continued public health neuroinfectious burden in Africa; it accounts for significant pro-
Received 22 January 2016 portion of lengthy intensive care unit (ICU) and hospital admissions.
Received in revised form 13 June 2016 Objectives: This study aimed to describe the pooled case-fatality rates of adult tetanus at African hospitals along
Accepted 11 July 2016 with relevant discussions and recommendations.
Available online 14 July 2016
Methods: A systematic review using advanced search strategies employing PubMed/MEDLINE and Web of Sci-
ence inclusive of gray literature handsearch was conducted for facility-based studies on adult tetanus by combin-
Keywords:
Tetanus
ing the terms “tetanus”, “Africa” spanning all previous years until January 15, 2016. PRISMA and MOOSE
Adult tetanus guidelines were followed. Studies from non-African countries and studies on neonatal and childhood tetanus
Africa were excluded. A meta-analysis with fixed- and random-effects model was performed to identify pooled mi-
Neuroinfection graine prevalence. Inter-study heterogeneity was analyzed employing I Oshinaike et al. (2012) (inconsistency).
Systematic review Results: Twenty-seven studies involving 3043 patients were included. Median age was 33.7 years (IQR 30–36).
Meta-analysis Median female to male ratio was 0.5. The geographic distribution of the studies was as follows: 15 (55.5%) studies
Case fatality were from Nigeria, 7 (26%) from Ethiopia, and the remaining single-centered studies were from Ghana (1; 3.7%),
Uganda (1; 3.7%), Senegal (1; 3.7%), Democratic Republic of Congo (1; 3.7%), and Tanzania (1; 3.7%). The majority
(88%) of the studies were from tertiary specialized or teaching university hospital settings.Median duration of the
study period was 6.5 years (IQR 4–9.25). Pooled crude tetanus case-fatality rate was found to be 43.2% (95% CI
36.9%–49.5%) on random-effects meta-analysis and 45.5% (95% CI 43.7%–47.2%) on fixed-effects meta-analysis.
There was considerable inter-study heterogeneity. A time-series observation did not reveal a trend of decreasing
case-fatality rates. Leading causes of death were complications from dysautonomia, aspiration pneumonia, hyp-
oxemia, and sepsis (in descending order). Longer incubation period and longer onset time were associated with
lower fatality; the further the wound site from the head, the longer the incubation period. Mechanical ventilation
was not available in 26% of the studies; where available, mechanical ventilation and ICU admission was not uti-
lized among most of the cases as the patients could not afford ICU care costs.
Conclusion: Despite declining tetanus incidence rates, case-fatality is still high in African care facilities. High rates
of tetanus case fatality indicate lower quality of medical care at hospital settings.Most common causes of death
are complication arising from dysautonomia and respiratory arrest secondary to laryngospasm. These can be
prevented by potent medications and mechanical ventilation; where resources are lacking, nursing in darker
and quieter rooms have been proven to be efficacious in reducing the frequency of spasms.
© 2016 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293

⁎ Corresponding author at: Advanced Clinical Consultation & Research Center, Addis Abäba, Ethiopia.
E-mail addresses: yohannes.woldeamanuel@gmail.com, ywoldeam@stanford.edu (Y.W. Woldeamanuel).

http://dx.doi.org/10.1016/j.jns.2016.07.025
0022-510X/© 2016 Elsevier B.V. All rights reserved.
 Y.W. Woldeamanuel et al. / Journal of the Neurological Sciences 368 (2016) 292–299 293

2.1. Inclusion and exclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294

2.2. Data extraction and pooled analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
2.3. Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Conflicts of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Appendix A. Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298

1. Introduction
Tetanus is a continued public health neuroinfectious burden in Afri-
ca; it accounts for significant proportion of lengthy intensive care unit
(ICU) and hospital admissions [1,2]. Agrarian lifestyle and tropical hot,
damp soil environment which is rich in organic matter - and hence con-
ducive for higher tetanus spore germination - are among the most im-
portant determinant risk factors for higher tetanus incidence in this
region [1,3]. Post-monsoon tropical seasons are associated with higher
incidence of tetanus because floods contaminate soil with Clostridium
spores [4]. Inadequate wound care at facilities and low awareness levels
at communities similarly contribute to the high incidence [1,2]. Aseptic
techniques of umbilical care increase the risk for contracting neonatal
tetanus which is known to be highly fatal [1,5].
Tetanus was once a scourge of Europe and North America; massive
rural to urban migration, improved public health, adequate protective
footwear, and education are considered to be the reasons that made tet-
anus become rare [1,3,6]. Tetanus burden had already started to show a
trend of significant decline from 1900s to 1940s prior to the introduc-
tion of tetanus toxoid and tetanus antitoxin immunization in the late
1940s [1,3]. Globally, disability-adjusted life years (DALYs) for tetanus
have decreased by almost two-third from 2000 to 2012; however, with-
in the same period in Africa, the reduction was lower than the global
rate by about 17% [7]. Similarly, the WHO African region (AFR) second
only to the South East Asian Region (SEAR) contributed the highest
shares of the global tetanus burden in 2005 (Fig. 1) [7,8].
Generalized tetanus has a high fatality rate [1,3]. Advanced medical
management has led to significant reduction of tetanus fatality to as
low as 13.2% [9]. In African hospitals and ICU settings, tetanus fatalities
remain high; this could be because of suboptimal care facilities with in-
adequate access to mechanical ventilation and potent medications, or
due to severe baseline illness secondary to lower healthcare access,
higher disease burden from remote rural locations, and lower disease
awareness [1]. Cost analyses of tetanus care indicate high economic bur-
den due to the natural course of tetanus which leads to protracted ward
or ICU admissions [10,11].
While longitudinal prospective studies provide the best evidence base
for data accrual, repeated cross-sectional studies can be regarded as
quasilongitudinal providing less costly but useful neuroepidemiologic
clues and outcome indicators for quality of medical care at facilities [12].
This study aimed to describe the pooled accumulating case-fatality rates
of adult tetanus at African hospitals along with relevant discussions and
recommendations. Time-series assessment of case-fatality rates was con-
ducted to observe the trend of hospital care quality.
2. Methods
The following six strategies were employed to capture our topic of
interest.
1. PubMed/MEDLINE® search was employed for Clinical Studies Cate-
gories on PubMed Clinical Queries tool combining the terms ‘tetanus
AND Africa’. Clinical Studies Category was selected for ‘Therapy’, ‘Eti-
ology’, ‘Diagnosis’, ‘Prognosis’, ‘Clinical Prediction Guides’, and scope
of search was made specific to ‘Broad’ to enable sensitivity and spec-
ificity search values of 99% and 70% [13], respectively.
2. PubMed/MEDLINE® search was employed for Clinical Studies Cate-
gories and Systematic Reviews on PubMed Clinical Queries Tool com-
bining the terms ‘Tetanus[Title] AND Africa[Title]’. Clinical Studies
Category was selected for ‘Therapy’, ‘Etiology’, ‘Diagnosis’, ‘Progno-
sis’, ‘Clinical Prediction Guides’, and scope of search was made specif-
ic to ‘Narrow’ to enable sensitivity and specificity search values of
93% and 97%13, respectively.
3. PubMed/MEDLINE search was employed without using the Clinical
Queries Tool. Search terms utilized were ‘Tetanus[Title/Abstract]
AND Africa[Title/Abstract]’.
4. Advanced PubMed/MEDLINE search was used by implementing
auto-suggested Medical Subject Headings (MeSH Terms) and Bool-
ean logic operator ‘AND’ as ‘(tetanus[MeSH Terms]) AND
Africa[MeSH Terms]’.
5. Web of Science Advanced Search was employed by using field tag ‘TS’
for topic, Boolean operator ‘AND’, and parentheses to create our
query as ‘TS = (tetanus AND Africa) on Indexes = CPCI (Conference
Proceedings Citation Index), Science Citation Index Expanded (SCI-EX-
PANDED), Timespan = All years. Results included all languages and
all document types.
6. Unpublished studies and relevant reference search was exhaustively
conducted using Google Scholar and ResearchGate. This enabled us
to capture unpublished studies from the gray literature (e.g. confer-
ence abstracts or research letters), with the goal of avoiding publica-
tion bias.

Fig. 1. Tetanus as percentage of total DALYs (Disability-adjusted Life Years) by WHO
region, 2005. The WHO African (AFR) and South East Asian Regions (SEAR) contribute
to most of the global tetanus burden. EUR: Europe; EMR: Eastern Mediterranean Region;
AMR: Region of the Americas; WPR: Western Pacific Region.
Two authors (YWW and ATA) reviewed each article, and where
disagreement occurred, discussion and consensus was achieved
with input from the each author. Methods were in accordance with
MOOSE [14] (Meta-analysis of Observational Studies in Epidemiology;
294 Y.W. Woldeamanuel et al. / Journal of the Neurological Sciences 368 (2016) 292–299
Supplementary 1) guidelines and PRISMA [15] (Preferred Reporting
Items for Systematic Reviews and Meta-analyses; Fig. 2) flow diagram.
All language publications were included, and French articles were trans-
lated. The search spanned all studies which are available in the medical
literature in previous years up to January 15, 2016. YWW is franco-
phone and expert on medical literature search strategies and has
previously contributed several related publications [16,17].
2.1. Inclusion and exclusion criteria
Studies reporting tetanus case-fatality rates on adult cohorts based
in African hospitals or ICU settings were included. Studies conducted
on children, and studies from non-African regions were excluded.
2.2. Data extraction and pooled analysis
Extracted data include study (First author and year of publication),
Country of Origin, Study Setting (Hospital, ICU), Female-to-Male Ratio,
Median/Mean age, and range (years), Total Number of Patients, Number
of Patients that Died, Duration of Study, and Case-Fatality of Adult Teta-
nus and 95% Confidence Intervals (CI).
2.3. Statistical analysis
Extracted data were pooled to combine adult tetanus case-fatality
data into one weighted magnitude. StatsDirect v2.7.9 (StatsDirect Ltd.,
Altrincham, Cheshire, UK) was used to analyze the results, develop
pooled prevalence rates and prepare forest plots. Combined pooled
Fig. 2. PRISMA (Preferred Reporting in Systematic Reviews and Meta-analysis) flow chart showing identification, screening, eligibility, and inclusion of the studies reviewed.
prevalence rates and 95% Confidence Intervals (CIs) were generated
using random-effects (DerSimonian-Laird) [18] and fixed-effects (in-
verse variance) meta-analysis model. Inter-study heterogeneity was an-
alyzed employing I2 (inconsistency). By virtue of being robust to outliers
and to nonparametric distribution, median and its interquartile ratio
(IQR) were selected to describe nominal data. Microsoft Excel
(Microsoft, Redmond, WA, USA) was utilized for data compilation and
computations.
3. Results
Twenty-seven studies [2,7,8,10,19–30] involving 3043 patients were
included (Table 1); the earliest study was conducted in 1958, and the
latest in 2014. Except one study which was published in French, all
the rest were in English. Median age was 33.7 years (IQR 30–36). Medi-
an female to male ratio was 0.5. The geographic distribution of the stud-
ies was as follows: 15 (55.5%) studies were from Nigeria, 7 (26%) from
Ethiopia, and the remaining single-centered studies were from Ghana
(1; 3.7%), Uganda (1; 3.7%), Senegal (1; 3.7%), Democratic Republic of
Congo (1; 3.7%), and Tanzania (1; 3.7%). The majority (88%) of the stud-
ies were from tertiary specialized or teaching university hospital set-
tings. Median duration of the study period was 6.5 years (IQR 4–9.25).
Pooled crude tetanus case-fatality rate was found to be 43.2% (95% CI
36.9%–49.5%) on random-effects meta-analysis (Fig. 3A) and 45.5% (95%
CI 43.7%–47.2%) on fixed-effects meta-analysis (Fig. 3B). There was con-
siderable inter-study heterogeneity with an I2 of 60%; the differences
among studies could be due to various factors including sample size
and quality of care. Time-series observation did not reveal a trend of
 Y.W. Woldeamanuel et al. / Journal of the Neurological Sciences 368 (2016) 292–299 295

Table 1
Description of the included studies.

Study [first author Country Study setting Female-to-male Median/mean Mechanical Total Number of Duration of study Case-fatality
and year of of origin (hospital, ICU) ratio age, and range ventilation (MV) number of patients that of adult
publication] (years) availability patients died tetanus

Johnstone, 1958 Nigeria University 3/4 N/A Not available 100 56 1954–1955 (1 year) 56%
hospital
Patel, 1970 Uganda University 1/4 N/A N/A 71 23 1966–1968 (2 years) 32.30%
hospital
Afonja, 1973 Nigeria University 1 Range = 11–79 Available 228 69 1967–1970 (3 years) 30.20%
hospital
Habte-Gabr, 1978 Ethiopia Teaching 5/8 Range = 20–66 Not available 27 14 1975–1976 (1 year) 51.85%
hospital
Hodes, 1990 Ethiopia 5 University 3/7 Mean age = 30; Not available 55 15 1983–1989 (6 years) 27.27%
hospitals range = 12–57
Oke, 2001 Nigeria University 1 1/9 Range = 10–75 Available 32 24 1990–1998 75%
hospital (8 years)
Hesse, 2003 Ghana Teaching 1/3 Mean age = 33; N/A; none received 158 79 1994–2001 (7 years) 50%
hospital range = 18–48 MV
(isolation unit)
Arogundade, 2004 Nigeria University 1/3 Mean age = 36; Not available 114 61 1992–1996 (4 years) 55%
hospital range = 19–54
Olubunmi, 2004 Nigreia University 1 3/7 Mean = 31.34; N/A; none received 66 18 1990–2000 (10 years) 26.20%
hospital range = 16–90 MV
Ojini, 2005 Nigeria University 1/2 Mean age = Available 349 129 1990–1999 (9 years) 36.90%
hospital 29.8; range =
10–88
Sounaré, 2005 Senegal Tertiary teaching N/A Mean age = 36 N/A 30 8 September–December 26.70%
hopsital 2002
(4 months)
Melaku, 2006 Ethiopia Tertiary 3/7 Mean age = 37; Available 146 72 N/A 49.20%
specialized range = 20–55
teaching hospital
Chapp-Jumbo, Nigeria University 3/5 Mean age = 31; Available 90 43 1993–2003 (10 years) 47.80%
2006 hospital range = N/A
Sanya, 2007 Nigeria University 4/9 Mean = 36.1 N/A 202 129 1990–2001 (11 years) 64%
hospital
Adudu, 2007 Nigeria University 3/8 N/A Available 26 16 1985–2003 (8 years) 52.10%
hospital
Komolafe, 2007 Nigeria University 3/7 Mean = 53; N/A; none received 79 36 1995–2004 (9 years) 45%
hospital range = 16–65 MV
Onwuekwe, 2008 Nigeria Tertiary hospital 5/7 Mean = 29.8; Available; but all 12 0 1999–2003 (4 years) 0
range = 18–59 managed at
medical wards
Ramos, 2008 Ethiopia Rural general 1/4 Median age = N/A; none received 5 1 2000–2005 (5 years) 20%
hospital 24; range = MV
15–27
Tadesse, 2009 Ethiopia Tertiary 1/2 Mean age = 35; Not available 29 12 2003–2008 (5 years) 41.40%
specialized range = 18–70
teaching hospital
Chukwubike, 2010 Nigeria University 3/4 Mean age = 30; Available 86 33 1996–2005 (9 years) 42.90%
hospital range = 17–44
Fawibe, 2010 Nigeria Tertiary 3/8 Mean age = 33; Available 35 20 2002–2006 (4 years) 57.10%
specialized range = 15–80
teaching hospital
Amare, 2011 Ethiopia Tertiary 1/3 Mean age = 33; Available 171 65 1996–2009 (13 years) 38%
specialized range N/A
teaching hospital
Muteya, 2012 DR Provincial N/A Mean age = 39; N/A 22 12 2005–2009 (4 years) 52.40%
Congo hospital range = 5–77
Oshinaike, 2012 Nigeria University 1/3 N/A Not available 176 99 2006–2011 (5 years) 56.20%
hospital
Amare, 2012 Ethiopia Tertiary 2/7 Mean age = 34; Available 68 24 2001–2009 (8 years) 35.30%
specialized range = 14–85
university
hospital
Bankole, 2012 Nigeria University 1/3 Mean age = 30; Available 190 31 2000–2009 (9 years) 16.30%
hospital range = 17–44
Melkert, 2014 Tanzania District hospital N/A N/A Not available 476 305 1962–2012 (50 years) 64%

decreasing case-fatality rates. The median time to presentation was variation by onset-to-admission. There was higher fatality among
found to be 3.6 days (IQR 2.7–5.5); there was only one study with out- cases that presented early to hospitals. Outcome data based on disease
come data available based on time of presentation. Based on that study, severity and incubation period (IP) revealed that higher severity and
we have prepared a bubble plot (Fig. 4) to demonstrate case-fatality longer IP were found to have increased mortality (Table 2). Case-fatality
296 Y.W. Woldeamanuel et al. / Journal of the Neurological Sciences 368 (2016) 292–299

was generally higher in patients with shorter OP (b48 h). The further
the wound site from the head, the longer the incubation period. Lower
limb injury was the most common portal of entry.
The highest case-fatality was 75%, while there was no fatality report-
ed in one study that included limited number of 12 cases [31]. The re-
maining fatality rates were heterogeneous. Most cases presented as
severe and generalized tetanus. Case-fatality of those patients with se-
vere and generalized tetanus were higher than moderate or mild and
cephalic or localized tetanus. Complications involving laryngospasm
and autonomic dysfunction such as cardiac dysrhythmia were associat-
ed with higher case-fatality. Other factors significantly associated with
higher case-fatality included older age (N40 years), male gender, pres-
ence of aspiration pneumonia, lower wound debridement, presence of
sepsis, higher sedative usage, and presence of abdominal rigidity. Lead-
ing causes of death were complications from dysautonomia, aspiration
pneumonia, hypoxemia, and sepsis (in descending order). Mechanical
ventilation was not available in 26% of the studies; where available, me-
chanical ventilation and ICU admission was not utilized among most of
the cases as the patients could not afford ICU care costs.
The majority of patients never received vaccination previous to ad-
mission. Tetanus toxoid vaccination was provided in about half of the
studies to those patients who survived on discharge. Tetanus immune
globulin was administered among few of the studies and it improved
survival. More information is provided in Supplementary 2.
4. Discussion
Pooled tetanus case-fatality rates were found to be high in Africa.
These fatality rates have not changed significantly compared to fatality
rates half a century ago. High case-fatality rates indicate either a highly
severe clinical form of tetanus in the region, highly virulent Clostridia
strains, severe forms of clinical presentation, or lower quality of medical
care at facilities. Inadequate drug supplies could lead to failure to man-
age complications associated with dysautonomia. Underutilization and/
or shortage of mechanical ventilation equipment need to be addressed.
Where there was no mechanical ventilation, laryngospasm-associated
respiratory failure was found to be the most common cause of death,
while tetanus-associated autonomic dysfunction was the leading

Fig. 3. A. Random-effects meta-analysis forest plot. Pooled crude tetanus case-fatality rate was found to be 43.2% (95% CI 36.9%–49.5%) on random-effects meta-analysis B. Fixed-effects
meta-analysis forest plot. Pooled crude tetanus case-fatality rate was found to be 45.5% (95% CI 43.7%–47.2%) on fixed-effects meta-analysis.
 Y.W. Woldeamanuel et al. / Journal of the Neurological Sciences 368 (2016) 292–299
Fig. 3 (continued).
cause of death in patients with ventilator support [1,32]. Many tetanus
patients presenting from rural settings could not afford ICU costs; be-
cause of this, they were managed in medical wards with insufficient
medical, surgical, and nursing care which results in higher fatality
rates. Some studies have shown the efficacy of managing and nursing
tetanus patients in dark quiet rooms to reduce the chances of precipitat-
ing spasms; this method can be feasible in rural hospitals which bear the
brunt of tetanus burden, and which have limited resources of ICU care
[1,31]. The absence of fatality in one the articles included within our
study similarly supports the possibility of minimizing tetanus fatality
rates utilizing low cost methods of nursing in dark quiet rooms [31].
The preponderance of male affliction could be explained by lower
outer-door risk for women in rural settings, and possibility of maternal
tetanus toxoid vaccination programs. Cost analysis were available from
only one study in Ethiopia published in 1978; the average per-patient
direct treatment cost during that period was ETB 159.92 (USD 79.96)
[10]. This cost analysis did not include indirect causes incurred by loss
of productivity related to tetanus ailment [10]. More recent cost analy-
ses from Brazil have shown that the direct per-patient treatment cost
could range from USD 257–977 depending on the severity of tetanus
297
grading [11]. In large population and rural settings of Africa facing lim-
ited resources, universal tetanus toxoid vaccination may not be a feasi-
ble and sustainable approach from a public health point of view;
additional high costs of vaccine cold-chain supply is a formidable oper-
ational and logistic challenge. According to the WHO [33], 11.2 million
infants in the following African and Asian regions have not received
their DTP3 vaccination: Democratic Republic of the Congo, South Africa,
Ethiopia, Nigeria, Uganda, India, Indonesia, Iraq, Pakistan, and the Phil-
ippines. Coverage of universal adult tetanus toxoid administration is
similarly low in most African countries [1]. Anti-toxin is known to be
available in some of the urban settings in Africa.
Poor compliance among traditionally entrenched community land-
scapes presents another challenge to conduct mass vaccination. Protec-
tive footwear and appropriate wound management are practicable and
more cost-efficient strategies which can reduce tetanus burden.
The higher fatality among cases that presented early to hospitals
could be because cases with severe and generalized tetanus spasms
seem to get more attention to be taken early for urgent hospital admis-
sion. The other reason could be the fact that poor quality of treatment
seen within these settings, even if intervened early might not result in
298 Y.W. Woldeamanuel et al. / Journal of the Neurological Sciences 368 (2016) 292–299

5. Conclusion

Despite declining tetanus incidence rates, case-fatality is still high in

African care facilities. High rates of tetanus case fatality indicate lower
quality of medical care at hospital settings.Most common causes of
death are complication arising from dysautonomia and respiratory ar-
rest secondary to laryngospasm. These can be prevented by potent
medications and mechanical ventilation; where resources are lacking,
nursing in darker and quieter rooms have been proven to be efficacious
in reducing the frequency of spasms.

Conflicts of interest

The authors declare that they have no conflict of interest.

Acknowledgements
Fig. 4. Bubble plots demonstrating Case-Fatality for Onset-to-Admission by Sample Size
(Johnstone et al. 1958, Nigeria). There was higher fatality among cases that presented
This research received no specific grant from any funding agency in
early to hospitals; this could be because cases with severe and generalized tetanus
spasms seem to get more attention to be taken early for urgent hospital admission. Blue the public, commercial, or not-for-profit sectors.
bubbles represent case-fatality; the bigger the size of the bubble, the higher the case-
fatality. Appendix A. Supplementary data

improving survival. That higher severity, longer IP, and higher OP were Supplementary data to this article can be found online at http://dx.
found to have increased mortality could indicate a low-cost locus for in- doi.org/10.1016/j.jns.2016.07.025.
tervention such as community education and awareness on early pre-
sentation. Period of onset (OP) is an important factor because it is References
associated with lower recall bias as the phenomenon of the spasms
[1] Y.W. Woldeamanuel, Tetanus in Ethiopia: unveiling the blight of an entirely vac-
strike the patient hence resulting in lower chances of being overlooked. cine-preventable disease, Curr. Neurol. Neurosci. Rep. 12 (6) (2012) 655–665.
It is a possibility that Africa's tetanus incidence may prospectively [2] O.O. Oshinaike, O.O. Ojelabi, A.O. Ogbera, O.O. Ojo, F.A. Ajose, N.U. Okubadejo, Im-
decrease in correlation with the current massive urbanization taking proving case fatality rate of adult tetanus in urban Nigeria: focus on better facilities
of care, Trop. Dr. 42 (4) (2012) 208–210.
place in the region. In general, poverty reduction and enhanced welfare [3] Centers for Disease Control and Prevention, Tetanus, http://www.cdc.gov/vaccines/
combined with universal healthcare access will undoubtedly eliminate pubs/pinkbook/downloads/tetanus.pdf2013.
the burden of tetanus both at facility and community level. [4] J. Bansal, J. Richa, G. K, Seasonal variation in non neonatal tetanus: 7 years retrospec-
tive record based study, Int. Multispecialty J. Health (2015) 1(4).
Table 2 [5] G. Tura, M. Fantahun, A. Worku, Neonatal care practice and factors affecting in
Time to presentation and related outcome among studies where data was available. The Southwest Ethiopia: a mixed methods study, BMC Int. Health Hum. Rights 15
median time to presentation was found to be 3.6 days (IQR 2.7–5.5); there was no out- (2015) 18.
[6] O. Simonsen, A.V. Bloch, I. Heron, Epidemiology of tetanus in Denmark 1920–1982,
come data available based on time of presentation. However, outcome data based on dis-
Scand. J. Infect. Dis. 19 (4) (1987) 437–444.
ease severity and incubation period (IP) revealed that higher severity and longer IP were
[7] D. Melkert, L. Kahema, P. Melkert, Reduction of mortality due to tetanus by
found to have increased mortality. immunisation and proper wound management of the patients in Sengerema Desig-
nated District Hospital, Tanzania, Trop. Dr. 44 (3) (2014) 163–165.
Studies Time from symptom onset to Did early presentation
[8] M.M. Muteya, A.K. Kabey, T.M. Lubanga, H.M. Tshamba, A.M. Nkoy, Prognosis of tet-
with data presentation improve outcome?
anus patients in the intensive care unit of Provincial Hospital Jason Sendwe, Lubum-
available
bashi, DR Congo, Pan Afr. Med. J. 14 (2013) 93.
Afonja, - 56% in first 2 days of symptom - No outcome data [9] Centers for Disease Control and Prevention, Tetanus surveillance - United States
1973. onset based on time to pre- 2001–2008, MMWR Morb. Mortal. Wkly Rep. 60 (2011) 365–369.
Nigeria - 26% in 3rd to 4th day after symp- sentation [10] E. Habte-Gabr, M. Mengistu, Tetanus in Gondar Public Health College Hospital, Ethi-
opia: a review of 72 cases, Ethiop. Med. J. 16 (2) (1978) 53–61.
tom onset - 73% of mortality had
[11] D.B. Miranda-Filho, R.A. Ximenes, N.T. Siqueira-Filha, A.C. Santos, Incremental costs
severe tetanus on ad-
of treating tetanus with intrathecal antitetanus immunoglobulin, Tropical Med. Int.
mission
Health 18 (5) (2013) 555–563.
- 20% moderately se- [12] D.J. Corsi, M. Neuman, J.E. Finlay, S.V. Subramanian, Demographic and health sur-
vere veys: a profile, Int. J. Epidemiol. 41 (6) (2012) 1602–1613.
- 7% mild [13] R.B. Haynes, K.A. McKibbon, N.L. Wilczynski, S.D. Walter, S.R. Werre, T. Hedges, Op-
Amare, - Mean and median duration of the - No outcome data timal search strategies for retrieving scientifically strong studies of treatment from
2012. initial symptom prior to presenta- based on time to pre- Medline: analytical survey, BMJ 330 (7501) (2005) 1179.
Ethiopia tion were 6.7 and 4 days, sentation [14] D.F. Stroup, J.A. Berlin, S.C. Morton, et al., Meta-analysis of observational studies in
respectively. - IP (incubation period) epidemiology: a proposal for reporting. Meta-analysis of observational studies in
N 10 days: dead, 7.7%; epidemiology (MOOSE) group, JAMA 283 (15) (2000) 2008–2012.
[15] D. Moher, A. Liberati, J. Tetzlaff, D.G. Altman, P. Group, Preferred reporting items for
alive, 92.3%
systematic reviews and meta-analyses: the PRISMA statement, PLoS Med. 6 (7)
- IP 7–10 days: dead
(2009), e1000097.
42%, alive 58.3%
[16] Y.W. Woldeamanuel, B. Girma, A 43-year systematic review and meta-analysis:
- IP b 7 days: dead 62%, case-fatality and risk of death among adults with tuberculous meningitis in Africa,
alive 38% J. Neurol. 261 (5) (2014) 851–865.
Hesse, - Time to presentation after the in- - No outcome data [17] Y.W. Woldeamanuel, A.P. Andreou, R.P. Cowan, Prevalence of migraine headache
2005. jury: 8.3 ± 22.4 weeks (range available and its weight on neurological burden in Africa: a 43-year systematic review and
Ghana 1–56 weeks) meta-analysis of community-based studies, J. Neurol. Sci. 342 (1–2) (2014) 1–15.
- Mean duration of symptoms to [18] R. DerSimonian, R. Kacker, Random-effects model for meta-analysis of clinical trials:
presentation: 2.9 ± 2.7 days an update, Control. Clin. Trials 28 (2) (2007) 105–114.
(range 1–21 days) [19] O.P. Adudu, O.A. Ogunrin, O.G. Adudu, Morbidity and mortality patterns among neu-
Tadesse, - Duration of illness at admission: - No outcome data rological patients in the intensive care unit of a tertiary health facility, Ann. Afr. Med.
2009. 1–15 days (mean 4.3 ± 3.7 days) available 6 (4) (2007) 174–179.
[20] A.O. Afonja, B.O. Jaiyeola, O.L. Tunwashe, Tetanus in Lagos: a review of 228 adult Ni-
Ethiopia
gerian patients, J. Trop. Med. Hyg. 76 (7) (1973) 171–174.
 Y.W. Woldeamanuel et al. / Journal of the Neurological Sciences 368 (2016) 292–299
[21] D.D. Johnstone, Tetanus in Nigeria; review of 100 cases treated in Ibadan, Br. Med. J.
1 (5061) (1958) 12–16.
[22] F.O. Ajose, O.O. Odusanya, Survival from adult tetanus in Lagos, Nigeria, Trop. Dr. 39
(1) (2009) 39–40.
[23] A. Amare, Y. Melkamu, D. Mekonnen, Tetanus in adults: clinical presentation, treat-
ment and predictors of mortality in a tertiary hospital in Ethiopia, J. Neurol. Sci. 317
(1–2) (2012) 62–65.
[24] A. Amare, A. Yami, Case-fatality of adult tetanus at Jimma University Teaching Hos-
pital, Southwest Ethiopia, Afr. Health Sci. 11 (1) (2011) 36–40.
[25] F.A. Arogundade, I.S. Bello, E.A. Kuteyi, A. Akinsola, Patterns of presentation and
mortality in tetanus: a 10-year retrospective review, Niger. Postgrad. Med. J. 11
(3) (2004) 198–202.
[26] E.O. Bandele, O.O. Akinyanju, B.J. Bojuwoye, An analysis of tetanus deaths in Lagos, J.
Natl. Med. Assoc. 83 (1) (1991) 55–58.
[27] I.A. Bankole, M.A. Danesi, O.O. Ojo, N.U. Okubadejo, F.I. Ojini, Characteristics and out-
come of tetanus in adolescent and adult patients admitted to the Lagos University
299
Teaching Hospital between 2000 and 2009, J. Neurol. Sci. 323 (1–2) (2012)
201–204.
[28] E.N. Chapp-Jumbo, Neurologic infections in a Nigerian university teaching hospital,
Afr. Health Sci. 6 (1) (2006) 55–58.
[29] O.A. Chukwubike, A.E. God'spower, A 10-year review of outcome of management of
tetanus in adults at a Nigerian tertiary hospital, Ann. Afr. Med. 8 (3) (2009)
168–172.
[30] A.E. Fawibe, The pattern and outcome of adult tetanus at a sub-urban tertiary hos-
pital in Nigeria, J. Coll. Physicians Surg. Pak. 20 (1) (2010) 68–70.
[31] I.O. Onwuekwe, C.C. Onyedum, A.C. Nwabueze, Experience with tetanus in a tertiary
hospital in south east Nigeria, Niger. J. Med. 17 (1) (2008) 50–52.
[32] C.L. Thwaites, L.M. Yen, H.T. Loan, et al., Magnesium sulphate for treatment of severe
tetanus: a randomised controlled trial, Lancet 368 (9545) (2006) 1436–1443.
[33] Immunization Coverage. Factsheet. (Reviewed March 2016). World Health Organi-
zation. (http://www.who.int/mediacentre/factsheets/fs378/en/ Accessed on June
1, 2016).