Content uploaded by Aamir A. Hamza
Author content
All content in this area was uploaded by Aamir A. Hamza on Jan 07, 2015
Content may be subject to copyright.
© Sudan JMS Vol. 6, No.3. Sept 2011 199
bbÜÜ||zz||ÇÇttÄÄTTÜÜàà||vvÄÄxx
The relation between the presence of intestinal bacteria in the perianal abscess
and the anticipated perianal fistula.
Saadeldin A. Idris*1, Aamir A. Hamza2, Isam MA Alegail (Late)3
Abstract:
Background: Perianal abscess and fistulas are frequently encountered in our surgical practice;
however, very little has been written about them. The clinical effectiveness of pus swabs for
microbiological analysis during incision and drainage of perianal abscess is controversial. Its cost
implication is often overlooked.
Objectives: To assess whether the culture of pus swab following incision and drainage of perianal
abscess has any significant impact on surgical outcome and on early anticipation of development of
fistula in-ano in our local population.
Patients and methods: A non-probability sample, total coverage multicenteric cross-sectional study.
All consecutive cases of incision and drainage of perianal abscess between November 2008 and
June 2011 were enrolled into our study after acceptance of a pre-given informed consent.
Results: Out of 76 patients with perianal abscess included in the study only 62 patients were
available for the final assessment (9 F: 53 M; age range: 18–63 years; mean ± SD: 37.66 ± 10.67).
Median follow-up was 2 months irrespective to culture result. The mean hospital stay was 1.44
days, and is affected by the presence of associated illness (p=0.02). Skin flora organisms, heavy
mixed growth of both skin flora and intestinal organisms, and no bacterial growth were isolated
from 75.8%, 12.9% and 11.3% respectively. Fistula in-ano developed in 16.7%, 83.3% and 0%
respectively. 83.3% of fistula developed in the group of patient who presented 10 days, which is
statistically found to be significant p=0.003. Fistula was developed in 7.1% (4/56) of patients who
were treated under general anaesthesia, and in 33.3% (2/6) of patients who were treated under local
anaesthesia.
Conclusions: The preliminary findings suggest that microbiological results have correlation with
presence of fistula in-ano. The result warranted us to submit pus swabs from perianal abscess for
assessment as it affect clinical effectiveness of treatment, and that culture of pus in perianal abscess
is an essential part of its management.
Keywords: Anal pain; Abscess; Fistula; Intestinal bacteria.
PERIANAL abscess (PAA) is a
generic term encompassing the
collection of pus to form an abscess
in the perianal, intersphincteric, ischiorectal
or pelvirectal spaces1. Most perianal
abscesses arise from the occluded duct of an
anal gland with subsequent bacterial
overgrowth and abscess formation2. Infection
usually caused by, an aerobic and anaerobic
polymicrobial infection. Bacteroides fragilis
is the predominant anaerobe.
1. Faculty of Medicine, Western Kordufan University,
Department of Surgery, Elnihood,Sudan.
2. Juba University, Department of Surgery, Khartoum,
Sudan.
3. Charity teaching hospital, Khartoum, Sudan.
*E-mail: saadeldinahmed@hotmail.com.
Other common bacteria include Escherichia
coli and those of the genera Proteus,
Bacteroides, and Streptococcus. Sources of
bacteria are skin, bowel, and, rarely, the
vagina3.
Perianal suppuration is common, affecting
men three times more frequently than
women1. PAAs are usually sporadic, but in
certain situations such as diabetes mellitus,
Crohn’s disease and chronic steroid treatment
the risk of developing abscess is increased.
The patient presents with acute anal pain and
tenderness. The pain often prevents digital
examination, and so an anaesthetic is required
for proper assessment. The diagnosis is
confirmed by demonstration of a localized
A
Saadeldin A. Idris et al. Intestinal bacteria in the perianal abscess and the anticipated fistula.
© Sudan JMS Vol. 6, No.3. Sept 2011 200
pea-sized lump in the intersphincteric space1.
An abscess should be incised and drained
under general anaesthetic4. Antibiotics have
little part to play, they are unlikely to abort
the infection once the symptoms have been
present for 24 hours, as they cannot penetrate
into the pus and there is often some necrosis
of fatty tissue4, 5. In other hand the routine use
of antibiotics following incision and drainage
of an abscess has no impact on healing time
or reduction of recurrence rates and therefore
not ordinarily indicated6-8. In addition, the
American Heart Association advises
preoperative antibiotics before incision and
drainage of infected tissue in patients with
prosthetic cardiac valves, previous bacterial
endocarditis, complex congenital heart
disease, surgically constructed systemic
pulmonary shunts or conduits, congenital
cardiac malformations, acquired valvular
dysfunction (e.g., rheumatic heart disease),
hypertrophic cardiomyopathy, and mitral
valve prolapse with valvular regurgitation
and/or thickened leaflets9.
Fistula-in-ano (FIA) is a chronic consequence
of PAA. Fifty percent of perianal abscesses
present with chronic purulent discharge or
intermittent pain associated with abscess
reaccumulation followed by intermittent
spontaneous discharge. Parks was classified
the FIA according to its relation to the anal
sphincter as intersphincteric, transsphincteric,
suprasphincteric, or extrasphincteric10.
The term “complex” fistula is frequently
encountered in literature, it denotes a fistula
when its track crosses >30 to 50 percent of
the external sphincter (high-transsphincteric,
suprasphincteric, and extrasphincteric. These
types of fistula their treatment poses a higher
risk for impairment of continence11-13.
There is relationship between perianal abscess
and fistula because suppuration is the
common aetiological factor for both. Frank
abscess precedes some cases of fistula and
inappropriate surgical drainage of perianal
abscess is responsible for a small but
significant proportion of fistulae1. Some
studies showed that approximately two thirds
of patients with rectal abscesses treated by
incision and drainage or by spontaneous
drainage will develop a chronic anal fistula14.
In patients with recurrent perianal abscesses,
a fistula tract may be identified. However,
there is no need to search routinely for a
fistula when draining straightforward perianal
abscesses because probing may actually
inadvertently cause a fistula1. If an associated
fistula suspected, a routine examination under
general anaesthetic 10 days later to look for it
should be arranged. Factors suggesting an
associated fistula include: a recurrent abscess,
growth of enteric rather than cutaneous
organism in the pus culture, or pus draining
from an internal opening seen on
proctoscopy4.
The study conducted to achieve the cost
effectiveness in term of early detection and
treatment of fistula in-ano (FIA), determine
the association between the presence of a
certain bacteria and the development of FIA.
Patients and methods:
The study conducted in the surgical units of
Charity and Omdurman teaching hospitals,
Khartoum, capital city of the Sudan with
more than 6 million inhabitants. All patients
with a first attack of perianal abscess and
absence of identifiable fistula who attended
the surgical emergency department in the
period from November 2008 to June 2011
were enrolled in our study after acceptance of
a pre-given informed consent. The
participants completed a questionnaire based
on British Medical Research Council
recommendations. It included personal data,
habits, predisposing factors and subject
disease history. The participants had
undergone general physical examination, and
then enrolled into appropriate investigations.
Clinical examination included assessment of
induration and scarring in the perianal region,
external and internal openings to exclude the
presence of fistula without probing the tract
iatrogenic or inadvertent creation of fistula
tract might occur. All patients were assessed
in the lithotomy or prone jackknife position.
Examinations performed under general
anaesthesia. Prior to collect pus samples, the
perianal skin was disinfected twice using
alcohol solution (spirit) firstly then 10%
Saadeldin A. Idris et al. Intestinal bacteria in the perianal abscess and the anticipated fistula.
© Sudan JMS Vol. 6, No.3. Sept 2011 201
povidone iodine applied to skin and kept in
situ till dry. Then after, pus was aspirated
from the abscess cavity using a 5-ml needle
syringe to prevent contamination by local skin
microflora. The samples were then sent for
routine aerobic and anaerobic culture
Abscess was treated by incision, drainage
(I&D), and de-roofing the cavity to prevent
premature closure of an abscess cavity and
the patients advised for daily dressing. Then
the patients were followed up in outpatient
referred clinic. The first visit was arranged 2
weeks postoperatively. During which
proctoscopic examination was carried out.
Results:
A total of 76 cases were identified in the
study period. None of the patients were
known to suffer from pre-existing
inflammatory bowel disease, exposure to
radiation therapy or malignancy which would
have predisposed them to the formation of
abscesses or fistulae. 14 patients were lost
from follow-up or did not provide the culture
result and all were excluded from the study
(compliance rate of pus swab sampling was
94.7%). Therefore 62 patients (53 males and
9 females) remained for the final assessment.
Male to female ratio was 5.9:1. Their age
ranged between 18 and 63 years old, with
mean ± SD of 37.66 ± 10.67 years.
Perianal pain was the main presenting
symptom as it was seen in 59 (95.2%)
patients. Duration of illness on presentation
ranged from 2 to 14 days with mean ± SD of
5.53 ± 3.4 days.
All abscesses were classified as perianal, and
there were no ischiorectal abscesses.
Eight patients had an associated illness along
with the perianal abscess. It includes diabetes
mellitus (DM), and use of steroid therapy
(due to bronchial asthma and empty sella
syndrome) in 5 (8.1%) and 3 (4.8%) patients
respectively.
Their hospital stay ranged from 0 to 5 days,
with mean ± SD of 1.44 ± 0.97 days, that is
affected by the presence of associated illness
(p=0.02).
Eighteen (29.03%) patients had received
antibiotics during the preoperative period,
which was self-administered or prescribed by
doctors. There was observed effect of
preoperative antibiotic on culture growth, but
statistically was not significant (P=0.068)
A total of 69 bacterial growths were isolated
from the abscesses, either as single cutaneous
microorganism species (41 cases (66.1%)) or
as mixed growth of both skin flora and
intestinal organisms (14 of individual abscess
(22.6%)).
The remaining 7 (11.3%) specimens did not
yield any bacterial growth on culture, whether
aerobic or anaerobic.
Anaerobic microorganisms were isolated
from 6 out of 62 patients (9.7%), and aerobic
organisms were found in 55 (88.7%).
The culture rates of anaerobic and aerobic
bacteria were determined to be 8.70% (6
(anaerobes) /69 (total cultured organism
(aerobes+ anaerobes))) and 91.30% (63
(aerobes) /69 (total cultured organism
(aerobes+ anaerobes))), respectively.
Cultured aerobic bacteria included mainly
Streptococcus spp. in 59.7% (37/62) with
culture rate of 58.7% (37/69), Staphylococcus
aureus in 29.03% (18/62) with culture rate of
26.09% (18/69), and Escherichia coli in
8.06% (5/62) with culture rate of 7.25%
(5/69).
The predominant anaerobes cultured, were
Bacteroides, seen in 4/62 (6.45%) patients,
with culture rate of 5.8% (4/69).
The most common infectious anaerobic and
aerobic microorganisms found were
Bacteroides and Streptococcus; 4/62 (6.45%)
with culture rate of 5.8% and 37/62 (59.7%)
with culture rate of 53.62% respectively.
Where classified into cutaneous and intestinal
bacteria, there were 55 cutaneous bacteria and
14 intestinal bacterial isolates with a culture
rate of 100% and 25.46% respectively. The
most common infectious microorganisms
were Streptococcus spp. (37/62 (59.7%)) and
Escherichia coli (5/62 (8.06%)), their culture
rates were 53.62% (37/69) and 7.25% (5/69)
respectively. The identity of the cultured
organisms is shown in Table 1 and 2.
Saadeldin A. Idris et al. Intestinal bacteria in the perianal abscess and the anticipated fistula.
© Sudan JMS Vol. 6, No.3. Sept 2011 202
Table 1: Microbiological analyses according to their classification into aerobic and anaerobic of
perianal abscesses in study group, Khartoum 2011 (n=62).
Organisms Isolated bacteria Culture rate (%) N=69
Aerobic 63 91.30
Escherichia coli 05 07.25
Klebsiella 01 01.45
Streptococcus spp. 37 from 55 abscesses
01
53.62
Citrobacter 01.45
Staphylococcus aureus 18 26.09
Salmonella enteric 01 01.45
Anaerobic 06 08.70
Bacteroides 04 05.80
Clostridium perfringens 01 from 6 abscesses 01.45
Fusobacterium 01 01.45
Total 69 100.00
Table 2: Microbiological analyses according to their classification into cutaneous and intestinal
bacteria in study group, Khartoum 2011 (n=62).
Organisms Isolated bacteria (%) N=55 Culture rate (%) N=69
Cutaneous bacteria 55 100 79.71
Streptococcus spp. 37 from 55 abscesses
18
67.27 53.62
Staphylococcus aureus 32.73 26.09
Intestinal bacteria 14 25.46 20.29
Escherichia coli 05 09.09 07.25
Klebsiella 01 01.82 01.45
Citrobacter 01 01.82 01.45
Salmonella enterica 01 from 8 abscesses 01.82 01.45
Bacteroides 04 07.27 05.80
Clostridium perfringens 01 01.82 01.45
Fusobacterium 01 01.82 01.45
Total 69 — 100.00
Fifty-six (90.3%) of abscess drainage were
performed under general anaesthesia and the
reminder 6 (9.7%) were carried out under
local anaesthesia.
Within follow-up period of 4 months (median
2 months) 56 (90.3%) of the abscesses had
healed leaving 6 (9.7%) cases with fistula in-
ano. Subsequently out of the healed
abscesses 7 patients presented with recurrent
abscess formation. These patients were
followed up for 4 months again and on
repeated proctoscpy none of them developed
fistula in ano. Their culture results were
mixed cutaneous bacteria.
The incidence of perianal fistula in male and
female was 5/53 (9.4%) and 1/9 (11.1%)
respectively. On univariate analysis, men did
not differ significantly from women, p=0.67.
There was correlation between the duration of
symptoms and the development of fistula, as
83.3% (in 5 out of 6 patients with fistula) of
fistula developed in the group of patient who
presented with symptoms for more than or
equal to 10 days of duration, which is
statistically found to be significant p=0.003.
None of patients with associated medical
illness had developed fistula in-ano.
Saadeldin A. Idris et al. Intestinal bacteria in the perianal abscess and the anticipated fistula.
© Sudan JMS Vol. 6, No.3. Sept 2011 203
Fistula was developed in 7.1% (4/56) of
patients who were treated under general
anaesthesia, and in 33.3% (2/6) of patients
who were treated under local anaesthetics.
The type of anaesthetic used during incision
and drainage of an abscess using tests of
Conditional Independence (Mantel-Haenszel)
was found to be statistically significant p=
0.006.
There were 16 (25.8%) smokers of either
cigarette or shisha (Narjeela) and 5 alcoholics
from both sexes. Among smokers, male
represented 15/53 and female represented 1/9.
While in other hand among alcoholics male
represented 5/53 and there was no alcoholic
female.
Using univariate and multivariate regression
local anaesthetics and delayed presentation
found to be statistically significant p value
were < 0.05. Table 3 shows the Cox
proportional hazard ratios for development of
fistula in ano by risk factors.
Table 3: Cox proportional hazard ratios for development of fistula in ano by risk factors in study
group, Khartoum 2011.
Risk factor Univariate Multivariate
Hazard ratio (CI) P value Hazard ratio (CI) P value
Male gender 0.53 (0.32-1.22) 0.21 0.73 (0.35-1.37) 0.58
Age < 45 1.98 (1.01-3.56) 0.09 1.52 (1.26-3.89) 0.06
Smoker 1.17 (0.53-2.97) 0.48 1.21 (0.43-2.59) 1.02
Perioperative antibiotics 1.33 (0.71-2.64) 0.68 1.61 (0.89-2.43) 0.25
Diabetics 2.53 (1.11-6.97) 0.07 2.75 (0.74-6.32) 3.01
Patients on Steroid therapy 3.65 (1.51-4.01) 0.98 3.81 (1.35-2.59) 1.73
Local anaesthetics 1.03 (0.91-2.79) < 0.05 1.64 (1.01-2.07) < 0.05
Delayed presentation 8.28 (5.97-10.97) < 0.05 19.03 (7.79-10.12) < 0.05
Five out of 8 (62.5%) patients with intestinal
bacteria were developed fistula, where only 1
out of 47 (2.1%) with pure cutaneous bacteria
developed fistula in ano (FIA).
Statistically speaking, using Fisher's exact test
there was significant effect of isolation of gut
microorganisms in perianal abscess (PAA) on
the development fistula-in-ano (FIA) in the
future compared with isolation of non-gut
microorganisms (p= .001), table 4.
Table 4: Relation between the type of bacteria and the development of perianal fistula in study
group, Khartoum 2011 (n=62).
Culture result Proctoscopic examination Total
No fistula Fistula
Cutaneous bacteria 46 (97.9%)
3 (37.5%)
7 (100%)
1 (2.1%)
5 (62.5%)
0 (0%)
47 (75.8%)
8 (12.9%)
7 (11.3%)
Mixed bacteria
No growth
Total 56 (90.3%) 6 (9.7%) 62 (100%)
Discussion:
Perianal abscesses (PAAs) are important not
only because they constitute a common
problem of the perianal area15, but also
because they mostly affect the most active
part of population, a fact that was confirmed
by this study as the mean age of our patients
was 37.66 years and most of them (68.2%)
were in the age range of 18-45 years.
Saadeldin A. Idris et al. Intestinal bacteria in the perianal abscess and the anticipated fistula.
© Sudan JMS Vol. 6, No.3. Sept 2011 204
This is less than in the study by Hamadani A.
et al16. where the overall mean age at
presentation was 43.6 (median 43; range 20-
88) years.
Most of our patients were males and male to
female ratio of 5.9:1 is well seen within the
general distribution of the illness among sexes
in literature worldwide17, 18. This might be
explained by the trend of our female patients
to prefer private hospitals to general hospitals,
where the study was performed, when the
operation involves a socially sensitive part of
the body. However, a study conducted at St.
Mark’s Hospital in London did not reveal any
significant differences in circulating sex
hormone levels between patients with fistula-
in-ano compared with matched controls19.
Mean duration of illness on presentation was
long 5.53 days (range 2-14 days), none had a
previous history of an abscess. Safwan et al.20
in their study in Basrah, Iraq, reported almost
similar result as a mean period of presenting
symptoms was 5.17 days (range 1-15 days),
whereas in study by ST Edino et al.21 in
Nigeria, the duration of symptoms was 3-10
days (mean 6.4 days). This relatively long
duration might be attributed to the generally
low level of health awareness among our
patients and their negligence towards early
medical consultation, or it might be due to
their higher pain threshold.
The mean hospital stay, on the other hand,
was short (1.44 days) stressing the fact that
this illness, especially if managed early, is
potentially easily curable. It seems that the
additional hospital stay was related to the
associated illnesses rather than the perianal
abscess itself. This is particularly clear
especially in diabetic patients in whom few
more days were spent to control blood sugar.
In study by Safwan et al.21, postoperative
hospital stay ranged from 1-4 days with a
mean of 1.17 days, and it was found that the
effect of associated illnesses on prolongation
of hospital stay was statistically significant.
Smoking has been previously implicated as a
risk factor in the development or exacerbation
of chronic inflammatory diseases affecting
the skin or its appendages16. Although there
are limited data on the effect of smoking on
the development of anal sepsis, a case-
controlled study in San Diego22 noted that a
history of recent smoking was a significant
risk factor for development of anal fistula and
abscess. It is interesting to note that the
incidence of smokers in our study was 25.8%
of either cigarette or shisha (Narjeela) (16
patients), which is lower than the estimated
incidence of 29 percent in the general
population of urban areas within the United
States22, and the 41 percent that estimated by
Hamadani A. et al.16.
Only 5/53 [5/62 (8.06%)] of our males and
none of the female patients consume alcohol,
whereas in study by Varut Lohsiriwat et al. 23
in Thiland, the incidence of alcohol
consumption among patients with perianal
fistula was 21/64 (32.8%), this difference
might be due to religious factors.
The main presenting symptom in this study
was a perianal pain that had been encountered
in 59 patients (95.2%). Similar result was
reported before20. Therefore, the clinician
should never attribute acute anal pain to
thrombosed internal haemorrhoids, anal
fissure or perianal cellulites as these entities
are less encountered when compared with
abscess and misdiagnosis may allow occult
anal sepsis to progress untreated.
Associated medical conditions identified in
12.9% of the patients, included diabetes
mellitus (DM) in five patients (8.1%) and 3
(4.8%) had a history of receiving steroids due
to bronchial asthma and empty sella
syndrome. There was no patient with a history
of inflammatory bowel disease (IBD) or had
irradiation therapy. The effect of associated
illnesses on prolongation of hospital stay was
statistically significant; Safwan et al.20 in their
study of 90 patients with PAA, found that 17
patients (18.9%) had an associated illness
along with PAA and also concluded similar
result.
The study showed that 18 patients (29%)
received antibiotics during preoperative
period, this was found to be higher than the
13.9% (5/36) reported elsewhere24.
Knowing the microbiology of PAA is the
mainstay for correct management25.
Saadeldin A. Idris et al. Intestinal bacteria in the perianal abscess and the anticipated fistula.
© Sudan JMS Vol. 6, No.3. Sept 2011 205
The compliance rate with the surgical practice
of pus swab sampling and reaching laboratory
in our study was 94.7%, this is better than that
reported by Edmund Leung et al. 26in united
kingdom (UK) where it was 83.8%.
The study revealed a total of 69 bacterial
growths are isolated from 55 abscesses, either
as single microorganism species (66.1%) or
as mixed growth (22.6%). This is in
agreement with Safwan et al.20. Geographical
difference and acclimatization my play a role
in the observed little difference rate of
bacterial growths.
The study revealed a bacterial growth rate
that is different from that of Edmund Leung et
al.26.This can be explained by preoperative
antibiotic usage observed, but it was
statistically not significant, p>0.05.
In this study the anaerobic and aerobic
microorganisms were isolated from 6 (10.9%)
and 55 (100%) abscesses with bacterial
growth respectively. This is different from the
87.5% and 100% anaerobic, and 90% and
95% aerobic growths reported in the
literature27,28.
The most common infectious anaerobic and
aerobic microorganisms found were
Bacteroids in 5.8% (4/69), and Streptococcus
in 53.62% (37/69). This is not going with the
dominant E.coli (62.5%) and Bacteroids
(55.6%) reported by others27, 28.
Where classified into cutaneous and intestinal
bacteria, our study found that the most
common infectious microorganisms were
Streptococcus spp. and E.coli. This is
comparable with the result obtained by
Safwan et al.20
About 6 (9.7%) of abscesses were carried out
under local anaesthesia. Use of local
anaesthetic in the current study was either due
to the general condition of the patient at
presentation that the application of general
anaesthetic might affect his life, or some of
the patients presented at time where the
anaesthesia team were busy in emergency
aseptic theatre.
Follow up of drained PAA is encouraged,
because the estimated acute abscess
recurrences and development of chronic
fistula in-ano in literature occur in 10% and
up to 50% of patients respectively16.
The duration of follow-up was varied in
literature and ranged from 1.5 to 36
months16,20,23-27,29,30,(table 5), while in our
study we assessed patient only at 4 months.
The median follow up period was 2 months.
During follow up period, 10 patients (13.2%)
were lost from follow up so they were
excluded from the study. Comparatively, in
study by Edmund Leung et al.26 5% of
patients were lost to follow up.
Table 5: Post-surgical follow-up periods found in literature, Khartoum 2011 (n=62)
Author Follow up period/ month Median follow up/month
Y. El-Dhuwaib et al. 29 01.5 —
Edmund Leung et al. 26 06 —
Grace 25 02 — 36 06
Serour et al. 24 03 — 18 10
Safwan et al. 20 12 —
Cheng et al 27 — 19
Varut Lohsiriwat et al. 23 10 — 53 30
Re!it Inceo"lu et al. 30 06 — 78 35
Hamadani A. et al. 16 — 38
Our study 04 02
Although failure to return to the hospital is
not evident that a fistula does not exist, it is
likely that a large proportion of the patients
lost to follow-up had complete healing and
therefore decided that they did not require
further follow-up as they stated when
Saadeldin A. Idris et al. Intestinal bacteria in the perianal abscess and the anticipated fistula.
© Sudan JMS Vol. 6, No.3. Sept 2011 206
contacted by phone. It must be kept in mind;
however, that 81.6% of cases were assessed
in referred clinic for healing, recurrence or
development of FIA. Once again, follow-up
was based on clinical findings and not the
results of the microbiological analysis only.
Healing time which defined as the period
from the date of operation to the date of
complete healing ranged between 9 to 37
days, with mean ± SD of 16.9 ± 5.4 days. In
25.8% (16 patients) wound found completely
healed in less than 14 days, while in the
reminder 74.2% (46 patients) wounds were
healed in 14 days. This is comparable with
the result obtained by others26,30.
During the follow up period, 7 (11.3%)
patients were presented with recurrent abscess
formation that necessitate I & D and swap
culture again that isolated mixed cutaneous
bacteria, and they were followed up for
another four weeks, where none had
developed FIA.
In these abscesses, the main concern is the
presence of a concomitant fistula. There are
few good follow up studies relating to the
incidence of recurrence of abscess or
development of fistula in-ano (FIA) after
drainage of an acute perianal abscess (PAA).
The high incidence of recurrent sepsis in our
study is supported by studies showing that in
any group of patients presenting with perianal
sepsis there is high incidence of previous
episodes of sepsis24, 31, 32. In studies by Re!it
Inceo"lu et al.30 and Safwan et al.20 no
recurrence was observed.
During follow up period the incidence of
fistula development following management of
an acute PAA recorded in our study is 9.7%
(6/62 patients). This is in agree with that
reported in literature (5.77%— 85%)20,33-35.
The presented lower rate of FIA may be
attributed to the fact that other studies did not
exclude acute abscess with coexisting fistula
or condition associated with high risk fistula
development (IBD, Irradiation, ....etc). It is
also possible that we underestimated the FIA
incidence as some patients with FIA might
not seek medical attention because all of the
fistulae diagnosed at follow-up were
unrecognized by the patients.
From male and female patients who presented
with PAA, the FIA was developed in 9.4%
and 11.1% respectively, and this observed
difference statistically was not significant
(p>0.05). This is in agreement with a study
conducted at St. Mark’s hospital in London20.
Our study showed that there is statistically
significant correlation between the duration of
symptoms and the development of fistula, as
83.3% of patients who developed fistula had
late presentation ( 10 days). This delay in
presentation is explaining the attitude of our
patients who do not seek medical advice until
late.
There was observed effect of preoperative
antibiotic on culture growth, but it was not
statistically significant.
In agreement with others16, the study revealed
that the use of preoperative antibiotics was
not associated with development of FIA.
The study found that 8.1% of patients had
history of diabetes mellitus (DM) and none of
them developed FIA. This is different from
the FIA that developed in 10% of diabetic
patients in some reports23.
Use of local anaesthetic is statistically
significant in development of FIA, as 33.3%
(2/6) of fistula developed in patients who
were treated under local anaesthesia. Some
investigators identified inadequate surgery
and spontaneous drainage as factors are
responsible for subsequent recurrence25.
Our study showed that 62.5% of patients with
intestinal bacteria developed fistula,
compared to only 2.1% with pure cutaneous
bacteria. Statistically speaking, there is
significant effect of isolation of gut
microorganisms from PAA on the
development of FIA in the future compared
with isolation of non-gut microorganisms.
Safwan et al.20 documented that the
progression into perianal or FIA was observed
only in 5.77% of patients whose swabs
yielded growth of gut microorganisms. On the
other hand, no perianal fistula was reported to
develop among those patients whose cultures
showed other types of bacterial growth or no
growth at all.
Our study also has supported the earlier
Saadeldin A. Idris et al. Intestinal bacteria in the perianal abscess and the anticipated fistula.
© Sudan JMS Vol. 6, No.3. Sept 2011 207
reports of Grace et al.36 and Whitehead et al.34
who independently suggested that the
microbiology of anorectal sepsis with an
associated fistula differed from that where no
fistula was demonstrated.
Conclusion:
The clinician should never attribute acute anal
pain to thrombosed internal haemorrhoids,
perianal cellulitis or anal fissure as these
entities are less frequent, and misdiagnosis
may allow occult anal sepsis to progress
untreated.
Based on the present study, the incidence of
FIA following incision and drainage of PAA
was 9.7%. A diabetic and non-diabetic patient
appeared to have a similar risk for fistula
development.
Our study provides additional support for the
presence of gut bacteria in swab culture is a
good indicative of presence or development
of fistula in-ano, so bacteriology is to be
routinely requested following incision and
drainage of perianal abscesses. Intestinal
bacteria should be looked for specifically to
try to and to identify accurately those patients
with a risk of developing fistulas.
The presence of intestinal bacteria in culture
would justify a follow-up by proctoscope if
no fistula was found at the initial examination
by a surgeon after the initial episode has
resolved.
All perianal abscesses should be drained
under general anaesthesia, as the use of local
anaesthesia might lead to an incomplete
evacuation and drainage of the abscess, which
is in turn, might progress into FIA.
Early presentation is important to avoid
development of FIA. Associated illnesses as
DM or steroid therapy prolong hospital stay.
References:
1. James Garden O, Andrew W. Bradbury and John
Forsythe. Principles and practice of surgery, 4th edition
2002, pp357-60.
2. Parks A. Pathogenesis and treatment of fistula-in-
ano. BMJ 1961;1:463–9.
3. Brook I, Frazier EH. The aerobic and anaerobic
bacteriology of perirectal abscesses. J Clin
Microbiol. Nov 1997;35(11):2974-6.
4. C Parchment Smith and C Hernon. Anal and
perianal conditions. MRCS System Modules: Essential
Revision Notes. 2nd edition 2001:517-47.
5. Michael M. Henry and Jeremy N. Thompson.
Clinical surgery, 1st edition 2001:363-78.
6. Llera JL, Levy RC. Treatment of cutaneous abscess:
a double-blind clinical study. Ann Emerg Med
1985;14: 15–9.
7. Stewart MP, Laing MR, Krukowski ZH. Treatment
of acute abscesses by incision, curettage and primary
suture without antibiotics: a controlled clinical trial. Br
J Surg 1985;72:66–7.
8. Macfie J, Harvey J. The treatment of acute
superficial abscesses: a prospective clinical trial. Br J
Surg 1977;64: 264–6.
9. Dajani AS, Taubert KA, Wilson W, et al. Prevention
of bacterial endocarditis. Recommendations by the
American Heart Association. Circulation 1997;96:
358–66.
10. Parks AG, Gordon PH, Hardcastle JD. A
classification of fistula-in-ano. Br J Surg 1976;
63:1–12.
11. Parks AG, Stitz RW. The treatment of high fistula-
in-ano. Dis Colon Rectum 1976;19:487–99.
12. Kodner IJ, Mazor A, Shemesh EI et al. Endorectal
advancement flap repair of rectovaginal and other
complicated anorectal fistulas. Surgery 1993;114:
682–90.
13. Mizrahi N, Wexner SD, Zmora O et al. Endorectal
advancement flap: are there predictors of failure? Dis
Colon Rectum 2002;45:1616–21.
14. Serour F, Gorenstein A. Characteristics of perianal
abscess and fistula-in-ano in healthy children. World J
Surg. March 2006;30(3):467-472.
15. Nomikos IN. Anorectal abscesses: need for
accurate anatomical localization of the disease. Clin.
Anat. 1997; 10(4): 239-44.
16. Hamadani A, Haigh PI, Liu IL et al. Who is at risk
for developing chronic anal fistula or recurrent anal
sepsis after initial perianal abscess? Dis Colon Rectum
2009; 52: 217-21.
17. Sagap, Feza H Remzi. Controversies in the
treatment of common anal problems. World J
Gastroenterol 2006 May 28;12(20): 3146-3154.
18. Wu CL. Experience on the treatment of acute
anorectal abscess with primary fistulotomy. Gaoxiong
Yi. Xue. Ke. Xue. Za. Zhi. 1990; 6(5): 218-23.
19. Lunniss PJ, Jenkins PJ, Besser GM et al. Gender
differences in incidence of idiopathic fistula-in-ano are
not explained by circulating sex hormones. Int J
Colorect Dis 1995;10:25-8.
20. Safwan A Taha and Mahmoud S Al-Haroon. The
cilnical and bacteriological profile of ano-rectal
abscess in basrah; a prospective study. Bas J Surg,
2006; 1321-7.
21. S.T. Edino, C.H. Ihezue. Management of Acute
Abscesses in Jos, Nigeria. Nigerian Journal of Surgical
Research, 2001;3(1):24-28.
22. Doescher MP, Jackson JE, Jerant A et al.
Prevalence and trends in smoking: a national rural
study. J Rural Health 2006;22:112-8.
23. Varut Lohsiriwat, Hariruk Yodying, and Darin
Lohsiriwat. Incidence and Factors Influencing the
Saadeldin A. Idris et al. Intestinal bacteria in the perianal abscess and the anticipated fistula.
© Sudan JMS Vol. 6, No.3. Sept 2011 208
Development of Fistula-In-Ano after Incision and
Drainage of Perianal abscesses. J Med Assoc Thai
2010;93(1):61-5.
24.Francis Serour, Eli Somekh and Arkadi Gorenstien.
Perianal Abscess and Fistula-In-Ano in Infants: A
Different Entity? Dis Colon Rectum 2005;48:359-364.
25. Grace RH. The management of acute anorectal
sepsis. Ann R Coll Surg Eng 1990; l72:160–162.
26. Edmund Leung, Kirsten Mc Ardle, Marcelino
Yazbek-Hanna. Pus Swabs in Incision and Drainage of
Perianal Abscesses: What Is the Point? World J Surg
2009; 33:2448–2451.
27. Sheng-Fang Cheng, and Wen-Sy Tsai.
Microbiological Analysis of Perianal Abscess and its
Treatment. J Soc Colon Rectal Surgeon (Taiwan) 2010;
37-42.
28. Eykyn SJ, Grace RH. The relevance of
microbiology in the management of anorectal sepsis.
Ann R Coll Surg Engl 1986; 68: 237–239.
29. Y. El-Dhuwaib and B.J. Ammori. Peranal abscess
due to Neisseria gonorrhoeae: An Unusual Case in the
Post-Antibiotic Era. Eur J Cli Microbiol Infect Dis
2003;22:422-432.
30. Re!it Inceo"lu and Rasim Gencosmano"lu.
Fistulotomy and drainage of deep postanal space
abscess in the treatment of posterior horseshoe fistula.
BMC Surgery 2003;3:10
31. Mark H. Whiteford. Perianal Abscess/Fistula
Disease. Clin Colon Rectal Surg 2007;20:102–109.
32. Niyogi A, Agarwal T, Broadhust J et al.
Management of Perianal Abscess and Fistula-in-ano in
Chidren. Eur J Pediatr Surg 2010;20:35-39.
33. Ezer SS, Ogÿuzkurt P, Ince E et al. Perianal
abscess and fistula-in-ano in children: Aetiology,
management and outcome. J. Paediatr. Child Health
2010; 46: 92–5.
34. Whitehead SM, Leach RD, Eykyn SJ et al. The
aetiology of perirectal sepsis. Br J Surg 1982;69:166-8.
35. Nicholls G, Heaton ND, Lewis AM. Use of
bacteriology in anorectal sepsis as an indicator of anal
fistula: experience in a distinct general hospital. J R
Soc Med 1990;83:625-6.
36. Grace RH, Harper IA, Thompson RG. Anorectal
sepsis: microbiology in relation to fistula-in-ano. Br J
Surg 1982; 69: 401-3.
