Oral Diseases (2001) 7, 150–154
 2001 Munksgaard All rights reserved 1354-523X/01
www.munksgaard.dk
Clinical Oral Medicine
Seasonal variation of acute necrotising ulcerative
gingivitis in South Africans
TM Arendorf1, B Bredekamp1, C-A Cloete1, K Joshipura2,3
1
Department of Oral Medicine & Periodontology, Faculty of Dentistry, University of the Western Cape and WHO Oral Health
Collaborating Centre, Cape Town, South Africa; 2Department of Oral Health Policy and Epidemiology at Harvard School of Dental
Medicine, MA, USA; 3Department of Epidemiology at Harvard School of Public Health, MA, USA
OBJECTIVES: To examine seasonal variations in the
presentation of acute necrotising ulcerative gingivitis
(ANUG) in a previously unstudied population in Cape
Town, South Africa.
DESIGN: A descriptive study of all presenting cases of
ANUG at a health center over 6 years; patients were
examined at one point in time.
SETTING: A clinic-based cross-sectional study was car-
ried out among patients presenting routinely to the oral
medicine clinic at the University of the Western Cape.
Patients at this clinic were mostly colored and black
people, and were of low socioeconomic status.
SUBJECTS AND METHODS: All patients presenting to
the periodontal clinic were examined during the period
from 15 March 1992 to 15 March 1998 and interviews
were conducted. A comprehensive clinical intra-oral and
peri-oral examination was carried out using a dental mir-
ror and blunt probe by three trained calibrated dental
examiners. The minimum criteria for inclusion as an
ANUG patient were painful, bleeding gums with irrevers-
ible destruction of the interdental papillae (cratering)
and marginal gingiva. Patients with multiple small ulcer-
ations not restricted to the gingiva, but including other
oral sites such as the tongue and buccal mucosa were
diagnosed as herpetic stomatitis. Other distinguishing
characteristics of ANUG included presence of a pseudo-
membrane, odor, a strange taste in the mouth, and hy-
persalivation. Swabs and biopsies taken to verify the diag-
nosis when necessary. If there was any doubt as to the
diagnosis of the lesion it was not included among the
ANUG cases
MAIN OUTCOME MEASURE: Presence of ANUG.
RESULTS: Among the 19 944 patients presenting for per-
iodontal treatment, 684 were found to have ANUG. The
ANUG patients consisted of 58% males; 73% of the
Correspondence: Dr K Joshipura, Department of Oral Health Policy &
Epidemiology, Harvard School of Dental Medicine, 188 Longwood
Avenue, Boston, MA 02115, USA. Tel: 001 617 432 1455, Fax: 001 617
432 0047, E-mail: kjoshipura얀hms.harvard.edu
Received 4 November 1999; revised 24 January 2000; accepted 22 Nov-
ember 2000
ANUG patients were between the ages of 5 to 12 years.
The majority of ANUG patients presented during the
summer (55.4%), 27.7% in autumn and 8.4% during both
winter and spring. Significant differences were evident
between the numbers seen during the different seasons.
CONCLUSIONS: This study demonstrated a significant
seasonal variation in the occurence of ANUG in Cape
Town, South Africa.
Oral Diseases (2001) 7, 150–154
Keywords: acute necrotising ulcerative gingivitis; seasonal
variation
Introduction
Acute necrotising ulcerative gingivitis (ANUG) has been
reported widely and is not uncommon in developing coun-
tries (Enwonwu, 1972; Taiwo, 1993). Classical diagnostic
features include ulceration and necrosis of the interdental
papillae, pain and spontaneous gingival bleeding.
Additional features that may be present are a pseudomem-
brane and a characteristic ‘fetor oris’. Malaise, fever and
lymphadenopathy may be variably seen (Murayama et al,
1994).
At a meeting in Paris (November, 1992) convened by
the World Health Organisation (WHO) to discuss ANUG
and its possible destructive sequela noma (cancrum oris),
alarming reports were received that these conditions may
be on the increase in Africa and possibly Asia and Latin
America (WHO, 1994). The prevalence in these areas is
well reviewed by Osuji (1990).
The consistent presence of specific bacteria viz. fusobac-
teria and spirochaetes has suggested the aetiology could be
explained in microbiological terms but no definitive cause
and effect association has been established (Loesche et al,
1982). Other contributory factors implicated include poor
oral hygiene, mouth breathing, smoking, stress, sepsis, mal-
nutrition, systemic diseases including hormonal imbalance
and alterations in lymphocyte and neutrophil function
(Goldhaber and Giddon, 1964; Taiwo, 1993; Murayama et
al, 1994).

Another factor occasionally mentioned is a possible
association with climatic conditions (Pedler and Radden,
1957; Osuji, 1990). No published information on ANUG
in South Africa could be found. The incidence of ANUG
was examined at the Oral Health Centre, Faculty of Dentis-
try, University of the Western Cape, Cape Town, and this
showed that more cases of ANUG presented during certain
times of the year. It seemed justifiable to analyse ANUG
cases which were treated over a 6-year period (the time
period for which the Faculty has been at its present site).
The aim of this paper is to evaluate the hypothesis that the
incidence of ANUG may be related to seasons, and to dis-
cuss possible explanations for the variation of ANUG
across seasons.
Materials and methods
During the period from 15 March 1992 to 15 March 1998
a clinic-based cross-sectional study was carried out at the
Department of Oral Medicine and Periodontology, Oral
Health Centre, Faculty of Dentistry, University of the West-
ern Cape, Cape Town, on patients presenting routinely to
the oral medicine clinic. Patients suspected of ANUG were
evaluated further for a more definite diagnosis.
The investigation included completing the Faculty ques-
tionnaire detailing medical and dental information. Other
characteristics noted were age, gender and date of attend-
ance. A comprehensive clinical intra-oral and peri-oral
examination was carried out using a dental mirror and blunt
probe while the patient was seated in a dental chair fitted
with a fluorescent light (WHO, 1991). Oral examinations
were carried out by three calibrated dental examiners
(TMA, BB, C-AC) trained in oral medicine and the identi-
fication and management of ANUG.
Criteria for the diagnosis of ANUG were developed
(Goldhaber and Giddon, 1964; Johnson and Engel, 1986;
American Academy of Periodontology, 1992; Murayama et
al, 1994). The minimum criteria for inclusion as an ANUG
patient were painful, bleeding gums with irreversible
destruction of the marginal gingiva and interdental papillae
(cratering) confined to the gingiva. Palmer and Floyd
(1996) described these as ‘the most reliable diagnostic fea-
tures of ANUG’. Presence of a pseudomembrane, odor, or
a strange taste in the mouth, and hypersalivation were other
characteristics of ANUG. Diagnosis of ANUG was based
on clinical appearance with smears, swabs and biopsies
taken to verify the diagnosis when necessary. The criteria
for ANUG described above, were distinct from the criteria
for herpetic stomatitis. Herpetic stomatitis was rare in com-
parison to ANUG. Patients with multiple small ulcerations
not restricted to the gingiva, but including other oral sites
such as the tongue and buccal mucosa was diagnosed as
herpetic stomatitis. ANUG was characterized by restriction
to gingiva and loss of interdental papillae. There were very
few cases that presented with both ANUG and herpetic
stomatitis.
Large extensive and destructive necrosis which had sig-
nificant gingival involvement but extended to other adjac-
ent sites including facial soft tissue in individuals who clini-
cally presented with notable signs of malnutrition and
marked debilitation were identified as NOMA.
Acute necrotising ulcerative gingivitis in South Africa
TM Arendorf et al
151
If there was any doubt as to the diagnosis of the lesion
(eg, non-specific gingivitis or ulceration or herpetic gingi-
vostomatitis, etc) it was not included among the ANUG
cases.
Statistical methods
Relative risks, 95% confidence intervals and P-values were
computed for comparing the occurrence of ANUG cases
across seasons in the specified area. Each season was com-
pared to every other season individually, comparing two
seasons at a time. The Poisson distribution was used (a
denominator is not known—persons at risk, but assumed
to be constant across the seasons, hence cancels out).
Results
During the 6-year period, 664 patients with ANUG were
treated at the Faculty, most were male (387, 58.2%). This
constituted 3.3% of the 19 944 patients who presented for
periodontal treatment to the department. Their ages ranged
from 3 to 48 years with peak occurrences at 7 and 10 years.
Most of our patients (484, 73.0%) were in the 5 to 12 years
of age range and came from socially disadvantaged local
communities. Only six cases (0.9%) of noma (Prabhu and
Praetorius, 1993) were seen (four males) and all of these
were between 2 and 6 years.
Figure 1 shows the variation in ANUG cases over a sin-
gle year. This annual pattern of variation across the seasons
was similar over all 6 years of the study. Over the whole
study period, the majority (368, 55.4%) of patients with
ANUG presented in the summer months with 184 (27.7%)
seen in autumn and only 56 (8.4%) attending during both
the winter and spring periods (Figure 2).
Significant differences were evident between the num-
bers seen during the different seasons. This was particularly
marked when summer and to a lesser extent autumn was
compared with winter and spring. These seasons are of
comparable length.
The summer season was significantly different from aut-
umn, winter and spring. The cases of ANUG occurred
about six times as frequently in the summer as in the spring
(or winter), relative risk of 6.57 (95% confidence interval
= 4.96, 8.70). The cases were twice as frequent in summer
as in autumn (95% CI = 1.68, 2.39). Comparing autumn to
winter or spring the RR was 3.29 (95% CI = 2.44, 4.43). All
seasonal differences were highly significant (P ⬍ 0.001).
Discussion
Comparisons with other epidemiological studies are diffi-
cult as criteria for the diagnosis of ANUG and patients seen
in other studies may not have been similar to those used
in the present study. Additionally, misdiagnosis in some
studies is possible as acute primary herpetic gingivostomat-
itis and other oral diseases have similar clinical features
(Carranza, 1984). This might contribute, among other vari-
ables, to epidemiological reports ranging from less than 2%
to 27% (Sheiham, 1966; Enwonwu, 1972). Our finding of
3.3% falls within this range. Differences in socioeconomic,
demographic, local (eg, oral hygiene status, smoking
habits) and systemic factors (eg, nutritional and HIV status)
Oral Diseases
 Acute necrotising ulcerative gingivitis in South Africa
TM Arendorf et al
152
Figure 1 Variation in ANUG lesions over a year
Figure 2 Seasonal variations in ANUG
would also vary between different communities and coun-
tries. The questionnaire included questions on smoking
habits and socioeconomic status; however we were not able
to use this information as the response rate and data were
poor as patients were often unwilling to share this infor-
mation. The patient population in our study consisted pre-
dominantly of colored people and some black people. The
patients were from a low socioeconomic status, and
included a high proportion of smokers. These patients were
not tested for HIV, hence their HIV status was unknown.
ANUG is mainly seen in socially disadvantaged children
in developing countries (Pindborg et al, 1966; Enwonwu,
1972; Prabhu and Praetorius, 1993) with young adults being
more commonly affected in developed countries (Pindborg,
1951; Giddon et al, 1963). The present study shows that in
this region of South Africa the former pattern is evident.
Among the children in the present study those below 12
years of age appear to be the high-risk group (representing
Oral Diseases
almost three-quarters of our sample) and almost 80% were
below 20 years of age. Thus, the condition appears to have
a predilection mainly for children and young adults in
South Africa.
Compared with the rest of Africa (Prabhu and Praetorius,
1993) noma is infrequently seen and we speculate that
some, as yet undetermined, additional contributing factor(s)
might be present elsewhere on the continent.
Falkler and his co-workers (1999) isolated Fusobacter-
ium necrophorum, a pathogen primarily associated with
animal diseases, from human noma lesions in a Nigerian
sample and they postulate that this may have animal trans-
mission implications. They further speculate that the
initiation of noma might require a special consortium of
microorganisms, particularly F. necrophorum and Prevo-
tella intermedium.
Contreras et al (1997) summarised the literature which
associated ANUG with human cytomegalovirus, measles
virus, herpes simplex virus and other unspecified viruses.
In their own study on 62 Nigerian children, they found a
significantly higher prevalence of the herpesviridae, parti-
cularly the human cytomegalovirus in ANUG lesions of
malnourished children than in control groups. Herpes lab-
ialis may appear after exposure to sunlight and as there is
a longer and more intense exposure in the South African
summer we speculate that our findings could be explained
by a sunlight exposure, herpes-labialis, ANUG sequence.
Interestingly, the few viral infections for which national
and regional information was available for periods during
our study were more frequently reported during the sum-
mer. These included measles, gastro-enteritis and viral
hepatitis (De Villiers and Geffen, 1998; Naicker, 1998).
Seasonal variation in the incidence of ANUG was first
reported by Stones in 1954 (cited and confirmed by Pedler

and Radden, 1957). These British authors showed that in
their series the lowest incidence was found during the sum-
mer months and was highest in the winter but they offered
no explanation for this phenomenon.
As early as 1944, Stammers speculated that ‘frequent
colds’ were contributory to local predisposing causes
including poor or non-existent oral care (Stammers, 1944).
Assuming colds occur more frequently in the British winter,
this association with climatic conditions could be note-
worthy. Herpes viruses may play a role here as well
(Contreras et al, 1997).
In 1960 Wood suggested that there was a marked
reduction of blood flow to the skin induced by a cool
environment and an hypothesis has been advanced (Wood,
1960; Kardachi and Clarke, 1974) that low temperatures
induce gingival vascular peripheral constriction ‘aug-
menting the circulatory effects of sepsis, stress and smok-
ing’. These factors might be relevant to the pathogenesis
of ANUG in countries where this condition is more fre-
quently seen in winter.
However, this would not adequately explain the high
incidence of ANUG in hotter regions such as Nigeria, Gam-
bia, Egypt, Asia and South America (Pindborg et al, 1966;
Malberger, 1967; Jiminez and Baer, 1975, Osuji, 1990;
Cutler et al, 1994). A Nigerian study of 58 cases of ANUG
seen over a 1-year period also reported an association with
the seasons (Osuji, 1990).
Although no statistically significant difference was evi-
dent, 72% of cases were seen in the rainy season and this
was 2.6 times greater than the number of patients who
presented in the dry season. History of a higher frequency
of reported recent febrile illness (n = 55) during the rainy
season was thought to be contributory.
One of the limitations of this study was the fact that
recent febrile illness was not always specifically recorded
in the early stages of this study. Therefore, this possible
variation could not be accurately determined.
In the present study most cases of ANUG were seen dur-
ing the summer with a progressive decline in autumn, win-
ter and spring. The summer figures were more than twice
as great as the autumn figures and almost 6.6 times greater
than for winter and spring. This could not be explained
by variations in patient attendance (which did not differ
markedly between seasons) or slight differences in the
length of the seasons.
Interestingly, a study conducted on patients who experi-
enced more than one attack of ANUG showed that they
had ‘inadequate thermoregulatory mechanisms and might
use the mouth for heat exchange’ (Clarke and Giddon,
1971). We speculate that mouth breathing (Stammers,
1944) which is probably more common in summer and also
dehydration might contribute to this marked preponderance
of ANUG in summer. No other obvious local inter-seasonal
differences are known for the well-documented risk factors
for ANUG. These would include emotional stress, smoking,
hormonal imbalance and nutritional deficiencies
(Murayama et al, 1994; Enwonwu et al, 1999). Perhaps
tooth brushing, eating and drinking patterns might change
during the summer holidays with an adverse effect on the
composition of the oral flora.
Health promotion and intervention strategies could be
Acute necrotising ulcerative gingivitis in South Africa
TM Arendorf et al
153
targeted seasonally to alert both the public and health prac-
titioners to the possible advent of ANUG and its potentially
disfiguring successor noma. This is important as ANUG has
a fairly similar clinical presentation to more than a dozen
other oral soft tissue diseases which would need different
management (Carranza, 1984).
Additionally, HIV seropositivity is becoming an increas-
ingly important suspected predictive variable for the pres-
ence of ANUG (Horning and Cohen, 1995).
Conclusions
Within the limitations of this study it may be concluded
that in Cape Town, South Africa:
(1) The presentation of ANUG demonstrates seasonal vari-
ation, being seen significantly more frequently in sum-
mer, somewhat less so in autumn and only occasionally
in winter and spring. It may be noteworthy that some
viral infections, eg measles, gastro-enteritis and viral
hepatitis are also reported more frequently in the sum-
mer.
(2) ANUG is usually seen in poor young children and
young adults.
(3) Noma only rarely accompanies ANUG and this might
be due to an absent co-variable in this area.
Acknowledgements
The authors gratefully acknowledge Professor CO Enwonwu for
expertise and guidance with this study and perusal of the final
drafts of the manuscript and Ms U Prins and Dr H Holmes for
valuable help in the preparation of this manuscript. This study
was supported by the Medical Research Council of South Africa.
References
American Academy of Periodontology (1992). Glossary of Per-
iodontal Terms. American Academy of Periodontology:
Chicago, p 38.
Carranza FA, Jr (1984). Acute gingival infections. In: Glickman’s
Clinical Periodontology 6th edn. WB Saunders: Philadelphia,
pp 146–157.
Clarke RE, Giddon DB (1971). Body geometry of patients who
had recurrent attacks of acute necrotising ulcerative gingivitis.
Arch Oral Biol 16: 205–213.
Contreras A, Falkler WA, Enwonwu CO et al (1997). Human
Herpesviridae in acute necrotizing ulcerative gingivitis in chil-
dren in Nigeria. Oral Microbiol Immunol 12: 259–265.
Cutler CW, Wasfy MO, Ghaffar K et al (1994). Impaired bacteri-
cidal activity of PMN from two brothers with necrotising ulcer-
ative gingivo-periodontitis. J Periodontol 65: 357–363.
De Villiers, Geffen LN (1998). The development of passive health
surveillance by a sentinel network of family practitioners in
South Africa. S Afri Med J 88: 256–259.
Enwonwu CO (1972). Epidemiological and biochemical studies
of necrotising ulcerative gingivitis and noma (cancrum oris) in
Nigerian children. Arch Oral Biol 17: 1357–1371.
Enwonwu CO, Falkler WA, Idigbe EO et al (1999). Pathogenesis
of cancrum oris (noma): confounding interactions of malnu-
trition with infection. Amer J Trop Med and Hyg 60: 223–232.
Falkler WA, Enwonwu CO, Idigbe EO (1999). Isolation of Fusob-
acterium necrophorum from noma (cancrum oris). Amer J Trop
Med and Hyg 60: 150–156.
Oral Diseases
 Acute necrotising ulcerative gingivitis in South Africa
TM Arendorf et al
154
Giddon DB, Goldhaber P, Dunning JM (1963). Prevalence of
reported cases of ANUG in a university population. J Periodon-
tol 34: 366–371.
Goldhaber P, Giddon DB (1964). Present concepts concerning the
etiology and treatment of acute necrotising ulcerative gingivitis.
Int Dent J 14: 468–496.
Horning GM, Cohen ME (1995). Necrotizing ulcerative gingivitis,
periodontitis and stomatitis. J Periodontol 66: 990–998.
Jiminez LM, Baer PM (1975). Necrotising ulcerative gingivitis in
children. A nine year clinical study. J Periodontol 46: 715–720.
Johnson BD, Engel D (1986). Acute necrotising gingivitis. A
review of diagnosis, etiology and treatment. J Periodontol 57:
141–150.
Kardachi BJR, Clarke NG (1974). Aetiology of acute necrotising
ulcerative gingivitis: a hypothetical explanation. J Periodontol
45: 830–832.
Loesche WJ, Syed SA, Langhon BE et al (1982). The bacteri-
ology of acute necrotising ulcerative gingivitis. J Periodontol
53: 223–230.
Malberger E (1967). Acute infectious oral necrosis among young
children in Gambia, West Africa. J Periodont Res 2: 154–162.
Murayama Y, Kurihara H, Nagai et al (1994). Acute necrotising
ulcerative gingivitis: risk factors involving host defense mech-
anisms. Periodontol 2000 6: 116–124.
Naicker M (1998). Department of Health Directorate: Health Sys-
tems Research and Epidemiology. Notifiable Medical Con-
ditions.
Osuji OO (1990). Necrotising ulcerative gingivitis and cancrum
oris (noma) in Ibandan, Nigeria. J Periodontol 61: 769–772.
Oral Diseases
Palmer RM, Floyd PD (1996). A Clinical Guide to Perio-
dontology. British Dental Association: London, pp 55–56.
Pedler JA, Radden BG (1957). Seasonal incidences of acute ulcer-
ative gingivitis. Dent Rec 8: 23.
Pindborg JJ, Bhat M, Devanath KR et al (1966). Occurrence of
acute necrotising gingivitis in South Indian children. J Per-
iodontol 37: 14–19.
Pindborg JJ (1951). Influence of service in Armed Forces on inci-
dence of gingivitis. J Am Dent Assoc 42: 517.
Prabhu SR, Praetorius F (1993). Mixed bacterial infections of oral
tissues: fusospirochaetal diseases. In: Prabhu SR, Wilson DF
Daftary DK et al, eds. Oral Diseases in the Tropics. Oxford
Medical Publications: Oxford, pp 260–270.
Sheiham A (1966). An epidemiological survey of acute ulcerative
gingivitis in Nigerians. Arch Oral Biol 11: 937–942.
Stammers AF (1944). Vincent’s infection: observations and con-
clusions regarding the aetiology and treatment of 1017 civilian
cases. Br Dent J 76: 147–155, 171–177, 205–209.
Taiwo JO (1993). Oral hygiene status and necrotising ulcerative
gingivitis in Nigerian children. J Periodontol 63: 1071–1074.
Wood IE (1960). Effect of smoking on the peripheral circulation
in relation to environmental temperature. Ann NY Acad Sci
90: 114.
World Health Organisation recording form for oral lesions:
WHO/CRH/OMA, 1991.
World Health Organisation (1994). World Health Day. Oral
Health Unit. WHD94.1.1994.