Mawalla et al.

BMC Surgery 2011, 11:21

http://www.biomedcentral.com/1471-2482/11/21

RESEARCH ARTICLE Open Access

Predictors of surgical site infections among

patients undergoing major surgery at Bugando
Medical Centre in Northwestern Tanzania
Brian Mawalla1, Stephen E Mshana2*, Phillipo L Chalya1, Can Imirzalioglu3 and William Mahalu1

Abstract
Background: Surgical site infection (SSI) continues to be a major source of morbidity and mortality in developing
countries despite recent advances in aseptic techniques. There is no baseline information regarding SSI in our
setting therefore it was necessary to conduct this study to establish the prevalence, pattern and predictors of
surgical site infection at Bugando Medical Centre Mwanza (BMC), Tanzania.
Methods: This was a cross-sectional prospective study involving all patients who underwent major surgery in
surgical wards between July 2009 and March 2010. After informed written consent for the study and HIV testing,
all patients who met inclusion criteria were consecutively enrolled into the study. Pre-operative, intra-operative and
post operative data were collected using standardized data collection form. Wound specimens were collected and
processed as per standard operative procedures; and susceptibility testing was done using disc diffusion technique.
Data were analyzed using SPSS software version 15 and STATA.
Results: Surgical site infection (SSI) was detected in 65 (26.0%) patients, of whom 56 (86.2%) and 9 (13.8%) had
superficial and deep SSI respectively. Among 65 patients with clinical SSI, 56(86.2%) had positive aerobic culture.
Staphylococcus aureus was the predominant organism 16/56 (28.6%); of which 3/16 (18.8%) were MRSA. This was
followed by Escherichia coli 14/56 (25%) and Klebsiella pneumoniae 10/56 (17.9%). Among the Escherichia coli and
Klebsiella pneumoniae isolates 9(64.3%) and 8(80%) were ESBL producers respectively. A total of 37/250 (14.8%)
patients were HIV positive with a mean CD4 count of 296 cells/ml. Using multivariate logistic regression analysis,
presence of pre-morbid illness (OR = 6.1), use of drain (OR = 15.3), use of iodine alone in skin preparation (OR =
17.6), duration of operation ≥ 3 hours (OR = 3.2) and cigarette smoking (OR = 9.6) significantly predicted surgical
site infection (SSI)
Conclusion: SSI is common among patients admitted in surgical wards at BMC and pre-morbid illness, use of
drain, iodine alone in skin preparation, prolonged duration of the operation and cigarette smoking were found to
predict SSI. Prevention strategies focusing on factors associated with SSI is necessary in order to reduce the rate of
SSI in our setting.

Background worldwide [2]. Globally, surgical site infection rates have

Surgical site infections (SSIs) have been reported to be been reported to range from 2.5% to 41.9% [3-9]. In the
one of the most common causes of nosocomial infec- United States, approximately 2% to 5% of the 16 million
tions; is accounting 20% to 25% of all nosocomial infec- patients undergoing surgical procedures each year have
tions worldwide [1]. SSIs have been responsible for the postoperative surgical site infections [10].
increasing cost; morbidity and mortality related to surgi- In Tanzania, surgical site infections are still one of the
cal operations and continue to be a major problem leading causes of morbidity and mortality among
patients undergoing major surgery. Previous studies
* Correspondence: [email protected]
2
conducted in a district and a tertiary hospital in Tanza-
Department of Microbiology/Immunology Weill Bugando University College
of Health Sciences, Mwanza, Tanzania
nia reported the surgical site infections rate of 24% and
Full list of author information is available at the end of the article 19.4% respectively [11,12]. Surgical site infection is one
© 2011 Mawalla et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
Mawalla et al. BMC Surgery 2011, 11:21 Page 2 of 7
http://www.biomedcentral.com/1471-2482/11/21

of the most common preventable complications follow- before being enrolled into the study. HIV test was done
ing major surgery at Bugando Medical Centre and using national algorithm of rapid test [13].
represents a significant burden in terms of patient mor- The patients were assessed preoperatively, intraopera-
bidity, mortality and hospital costs. tively and postoperatively. Details that were recorded
Despite improvements in operating room practices, included; type of surgery, wound class, type and dura-
instrument sterilization methods, better surgical techni- tion of operation, antimicrobial prophylaxis, use of
que and the best efforts of infection prevention strate- drain, preoperative hospital stay and total hospital stay.
gies, surgical site infections remain a major cause of Each patient was followed up from the time of admis-
hospital-acquired infections and rates are increasing sion until time of the discharge and 30 days postopera-
globally even in hospitals with most modern facilities tively. Surgical wound was inspected at the time of the
and standard protocols of preoperative preparation and first dressing and weekly thereafter for 30 days. Superfi-
antibiotic prophylaxis. Moreover, in developing coun- cial surgical site infection was diagnosed if any one of
tries where resources are limited, even basic life-saving the following criteria was fulfilled: purulent drainage
operations, such as appendectomies and cesarean sec- from the superficial incision, organisms isolated from an
tions, are associated with high infection rates and mor- aseptically obtained culture of fluid or tissue from the
tality [4,10]. superficial incision, at least one of the following signs or
This study was conducted to establish the prevalence, symptoms of infection: pain or tenderness, localized
patterns and predictors of surgical site infection in our swelling, redness, or heat, and superficial incision is
local setting. Here we document multiple factors pre- deliberately opened by surgeon, and is culture-positive
dicting SSI in a large tertiary hospital in Mwanza, or not cultured. Deep surgical site infection was diag-
Northwestern, Tanzania. nosed if any one of the following criteria was fulfilled;
purulent drainage from the deep incision but not from
Methods the organ/space component of the surgical site, a deep
Study design incision spontaneously dehisces or is deliberately opened
This was an analytical cross-sectional prospective study by a surgeon and is culture-positive or not cultured and
among patients undergoing major surgery at BMC over the patient has at least one of the following signs or
a period of 9 months from July 2009 to March 2010. symptoms: fever (> 38°C), or localized pain or tender-
The study was conducted in the surgical wards of ness. A culture-negative finding does not meet this cri-
Bugando Medical Centre. Bugando Medical Centre terion, an abscess or other evidence of infection
(BMC) is a tertiary hospital with bed capacity of 900 involving the deep incision is found on direct examina-
and serves about 13 million people. tion, during reoperation, or by histopathological or radi-
ological examination, diagnosis of a deep incisional SSI
Sampling by a surgeon or attending physician. http://www.cdc.
Sample size was obtained using Kish and Lisle method, gov/nhsn/PDFs/pscManual/9pscSSIcurrent.pdf.
using an incidence of 19.4% from a previous study and Pus or pus swabs were obtained from surgical incision
the infection rate among wound classes and the use of and transported to the laboratory within an hour of col-
drain were also considered [12], the sample size lection. In the laboratory, the specimens were registered
obtained was 250. All patients of all age groups and gen- in the log books and processed as per standard operative
der undergoing major surgical procedures with visible procedures. Bacterial identification was done using an in
incision (laparotomy, excisional biopsy, appendicectomy, house biochemical panel [14,15]. Antibacterial suscept-
thyroidectomy, herniotomy, mastectomy, amputations, ibility testing to various antibiotics was performed using
open prostatectomy, cholecystectomy, thoracotomy, disc diffusion methods as previously described [16]. In
splenectomy etc) at Bugando Medical Centre who con- addition, blood was taken from all patients for random
sented for the study were serially recruited until the blood sugar testing and CD4 enumeration in HIV posi-
sample size was reached. The study was approved by tive status.
BMC/WBUCHS ethics review board.
Data analysis
Data collection and Laboratory procedures Data were entered into a computer using SPSS software
Predictor variables such as patient characteristics, preo- version 15 and analyzed using STATA software 10
perative data, intra-operative data and postoperative according to the objectives of the study. Chi- square test
data were obtained using a standardized data collection was used to determine for the significance associations
form. All patients who underwent major surgery with between the predictor and outcome variables to all cate-
visible incisions were eligible to participate in the study gorical variables and odds ratios were calculated to test
and requested to consent for study and HIV-testing for the strength of association between predictor
Mawalla et al. BMC Surgery 2011, 11:21 Page 3 of 7
http://www.biomedcentral.com/1471-2482/11/21

variables. In each variable reference variable was labeled excisional biopsy (23), thyroidectomy (21), appendicect-
as 1 in case of premorbidity each variable was compared omy (21) and others. The time taken for SSIs to develop
to all others. Significance was defined as a p-value of ranged from 3-12 days with a mean of 6.23 days. The
less than 0.05. In addition to univariate analysis multi- mean age was 38 years with standard deviation of 22.12
variate logistic regression analysis was performed. years. There were 116 (46.4%) males and 134 (53.6%)
females (Table 2). SSI rate between male and female
Study limitations was 58% and 40% respectively (p-value = 0.01).
Failure to perform anaerobic culture might have con-
tributed to the low prevalence of SSI due to inability to Predictors of SSI
establish presence of organisms that require such envir- Pre-morbid illness
onment (anaerobic bacteria). The absence of antibiotic Fifty-seven (22.8%) patients had pre-morbid illness
policy in the surgical wards might have influenced the namely diabetes Mellitus 18 (7.2%), hypertension 37
results of this study. Other surgical departments such (14.8%) and HIV 37 (14.8%). The SSI rates for patients
orthopedic and gynecological departments were not with pre-morbidity and those without were 70.2% and
included. 38.4% respectively (p-value = 0.002) (Table 2). Mean CD
4 count was 296.35 cells/μl, the rate of SSI among HIV
Results patients with CD count below 200 cells/μl and those
Study population and demographic characteristics with CD4 count of 200 cells/μl and above were 56.3%
A total number of 941 patients underwent major sur- and 18, 8% respectively (p-value = 0.0001).
gery at Bugando Medical Centre during the study per- History of cigarette smoking
iod. Two-hundred sixty five patients fulfilled the In this study, 33 (13.2%) patients had history of cigarette
inclusion criteria of these, 15 patients were excluded smoking. Of these 28 patients (84.8%) developed SSI.
from the analysis due to loss of follow up Surgical site There was a statistically significant association between
infection (SSI) was detected in 65 patients, giving an cigarette smoking and SSI (p-value = 0.001) as seen in
overall infection rate of 26.0% (table 1), of which 56 Table 2.
(86.2%) were superficial SSIs and 9(13.8%) were deep Antimicrobial prophylaxis
SSIs. Various procedures as seen in table 1 were done, Forty-one patients (16.4%) received preoperative antimi-
laparotomy (61) due to various reasons such as peritoni- crobial prophylaxis. Majority of the patients 27(65%)
tis, intestinal obstruction, intestinal perforation was the
leading procedure followed by open prostatectomy (26),
Table 2 Patients factors associated with Surgical Site
Infection (SSI) at Bugando Medical Centre
Table 1 Surgical procedures and SSI rate in each Variable SSI
procedure
Yes n (%) No n (%) P value OR 95% CI
Surgical Procedure SSI Positive SSI negative Total N = 65 N = 185
Laparotomy 17 (27.9%) 44 (72.1%) 61 Age in years
Open prostatectomy 8 (30.8%) 18 (69.8%) 26 < 21 12 (18.5%) 49 (26.4%) 1
Excisional Biopsy 7 (30.4%) 16 (69.4%) 23 21-40 18 (27.6%) 59 (32.9%) 0.60 1.24 0.54-2.83
Thyroidectomy 1 (4.8%) 20 (95.2%) 21 41-60 21 (32.3%) 47 (25.4%) 0.15 1.82 0.81-4.11
Appendicectomy 3 (15%) 17 (85%) 20 > 61 14 (21.5%) 30 (16.2%) 0.16 1.91 0.78-4.66
Herniotomy 0 (0.0%) 14(100%) 14 Sex
Spinal Bifida Repair 6 (46.2%) 7 (53.8%) 13 Female 26 (40%) 108 (58.3%) 1
Mastectomy 5 (45.5%) 6 (55.5%) 11 Male 39 (60%) 77 (42.7%) 0.01 2.12 1.19-3.70
Mayors Repair 3 (27.3%) 8 (62.7%) 11 Pre-morbidity
Amputation 1 (12.5%) 7 (87.5%) 8 None 25(38.5%) 167 (90.3) < 0.001 0.072 0.03-0.17
Cholecystectomy 1 (14.3%) 6 (85.7%) 7 Diabetes 9 (13.8%) 1 (0.5%) 0.001 29.6 4.3-281
Thoracotomy 4 (66.7%) 2 (33.3%) 6 Hypertension 15(23.1%) 13 (7.0%) 0.002 4.0 2.1-10.8
Splenectomy 1 (33.3%) 2 (67.7%) 3 HIV/AIDS 7 (10.7%) 2 (1.0%) < 0.001 11.0 2.6-64.2
Cystolithotomy 1 (33.3%) 2 (67.7%) 3 Pre-morbidity
Skull elevation 0 (0.0%) 3 (100%) 3 Absent 26 (40%) 168 (90.8%) 1
Others (20) 7 (35%) 13 (65%) 20 Present 39 (60%) 17 (9.2%) < 0.001 14.8 7.3-29.95
Total 65 (26%) 185 (74%) 250 Smoking
Others (Bronchotomy (1), Cleft lip repair (2), Contracture release (1), corrective No 37 (56.9%) 180 (97.3%) 1
osteotomy (2), mamoplasty (1), penis amputation (2), pinnaplasty (2), Stripping Yes 28 (43.1%) 5 (2.7%) < 0.001 27 9.87-75.2
(1), Urethroplasty (2), Varicolectomy (1), Z-plasty (2)
Mawalla et al. BMC Surgery 2011, 11:21 Page 4 of 7
http://www.biomedcentral.com/1471-2482/11/21

received ceftriaxone and metronidazole; ampicloxacillin respectively. The SSI rate in patient with duration of
and gentamicin 3 (7.3%); ceftriaxone alone 3 (7.3%); cef- operation < 3 hours was 20.9% and 50% in those which
triaxone and gentamicin 3 (7.3%); cloxacillin 3 (7.3%) had duration of operation ≥ 3 hours. There was statisti-
and ampicloxacillin and metronidazole 2 (4.9%). Anti- cally significant association between the duration of
biotics prophylaxis was administered in the following operation and SSI (p-value = 0.0001) (Table 3).
procedures; laparatomy (19), appendectomy (10), ampu- Post-operative antibiotic prophylaxis
tation (4), thoracotomy (2), mayo repairs (1), splenect- All, except three patients who underwent excisional
omy (1), excisional biopsy (1), spinal bifida repair (1) biopsy, were treated with antibiotics after the surgical
and bronchotomy (1). The timing of preoperative anti- operation. The frequency of post-operative antibiotics
microbial prophylaxis ranged from 1 hour to 5 hours prescription were; ampicloxacillin (24.3%), ceftriaxone
with the mean of 1.68 hours and standard deviation of -metronidazole (23%), ceftriaxone (9.7%), ampicloxacillin
0.96 hour. The rate of SSI among patients who received -gentamicin (8.5%), ampicloxacillin -metronidazole (8%),
pre-operative antimicrobial prophylaxis and those who gentamicin (7.2%), ceftriaxone-ampicloxacillin (6.1%),
did not receive were 14.63% and 28.23% respectively (p- ciprofloxacin (4.8%), chloramphenicol (4.8%) ampicloxa-
value = 0.07) Table 3. cillin-metronidazole-gentamicin (3.2%).
American Society of Anesthesiologists (ASA) classification, Surgical wound classification
skin preparation, use of drain Among the 250 patients, 193 (77.2%) had clean sur-
The SSI rates for ASA classification I, II and III were geries, 46 (18.4%) had clean-contaminated surgeries and
15.2%, 62.8% and 88.9% respectively (p-value = 0.001) 11 (4.4%) had contaminated surgeries. There were no
Table 2. The majority of patients 185 (74.0%) used a cases of dirty surgery. The rate of SSI was 21.24%,
combination of iodine and spirit for skin preparation. 43.48% and 36.36% in clean, clean -contaminated and
The SSI rates for iodine alone and Iodine + Spirit were contaminated wounds respectively (Table 4).
70.4% and 14.6%% respectively (p = 0.0001) Table 3. In Multivariate logistic regression
this study, the use of drain was recorded in 54 patients On multivariate logistic regression the following fac-
(21.6%). SSI rates among patients who used drain and tors were significantly found to predict SSI; pre-morbid
those who did not use were 48.2% and 19.9% respec- illness, duration of operation, use of iodine alone for
tively (p = 0.0001) Table 3. skin preparation, use of drain and cigarette smoking
Duration of operation (Table 5).
The duration of operation ranged from 40 minutes to 6 Bacterial infection and susceptibility pattern
hours with a mean of 1.9 and standard deviation of Out of 65 patients diagnosed to have SSI and specimens
0.83. Mode and median were 2.0 and 2.0 hours collected for microbiological investigation, 56 (86.2%)

Table 3 Pre operative factors associated with surgical site infection at Bugando medical centre
Variable Surgical site infection
Yes n (%), No n (%), P value OR 95% CI
N = 65 N = 185
Preoperative duration of hospitalization
≤ 7 days 41 (63%) 140 (75.7%) 1
> 7 days 24 (37%) 45 (24.3%) 0.05 1.82 0.99-3.33
Antimicrobial prophylaxis
Yes 6 (9.2%) 35 (18.9%) 1
No 59 (90.8%) 150 (81.1%) 0.08 2.29 0.91-5.73
Hair removal
Yes 1 (1.5%) 5 (2.7%) 0.59 1.78 0.20-15.5
No 64 (98.5%) 180 (97.3%) 1
ASA
I 30 (46.2%) 168 ((90.8%) 1
II 27 (41.5%) 16 (8.6%) < 0.001 9.45 4.55-16.6
III 8 (12.3%) 1 (0.6%) < 0.001 44.8 5.4-371.3
Skin preparation
Iodine alone 38 (58.5%) 16(8.7%) < 0.001 13.9 6.8-28.2
Iodine and spirit 27 (41.5%) 158 (86.5%) 1
ASA: American Society of Anesthesiologists
Mawalla et al. BMC Surgery 2011, 11:21 Page 5 of 7
http://www.biomedcentral.com/1471-2482/11/21

Table 4 Intra operative factors associated with surgical Table 5 Multivariate logistic regression analysis of factors
site infection at Bugando medical centre associated with SSI
Variable SSI Predictor variable Odd ratio 95% CI P- value
Yes n (%) No n (%) P OR 95% CI Pre-morbid illness 6.1 1.3-28.9 0.022
N = 65 N = 185 value ASA Classification above I 0.7 0.15-3.7 0.713
Nature of operation Use of drain 3.5 1.4-19.3 0.016
Elective 54 (83%) 142 1 Use of iodine alone in skin preparation 17.6 6.5-48.1 < 0.001
(76.7%)
Duration of Operation ≥ 3 hrs 3.2 1.1-9.3 0.033
Emergency 11 (17%) 43 (23.3%) 0.29 0.67 0.32-1.39
Use of Vicryl 2.4 0.78-7.2 0.129
Type of anaesthesia Cigarette smoking 9.6 2.4-38.3 0.001
GA 56 (86.1%) 161 (87%) 1
SAB 9 (13.9%) 24 (13%) 0.86 1.07 0.47-2.46
Surgeon prostatectomy. MRSA was detected in 3(19%) of Staphy-
Registrar 7 (10.7%) 32 (17.3%) 1 lococcus aureus; of 3 MRSA isolates 2 were isolated
Resident 13 (20%) 37 (20%) 0.37 1.6 0.57-4.5 from patients who had appendicectomy and one from
Specialist 45 (69.2%) 116 0.21 1.8 0.73-4.3 corrective osteotomy. The resistance rates to ciprofloxa-
(62,7%) cin were 86%, 80% and 54% for Escherichia coli, Kleb-
Use of drain siella pneumoniae and Staphylococcus aureus
Yes 26 (40%) 28 (15.1%) < 0.001 3.8 2.0-7.14 respectively.
No 39 (60%) 157 1
(84.9%) Discussion
Wound classification In this study, most of our patients were in the age group
Clean 41(63%) 152 (82%) 1 31-40 and showed a female preponderance. Similar
Clean contaminated 20 (30.7%) 26 (14%) 0.002 2.85 1.44-5.61 demographic observation was reported by another study
Contaminated 4 (6.3%) 7 (4%) 0.25 2.12 0.59-7.59 in India [17]. The rate of SSI was significantly higher in
Suture male patients than in females. This could be explained
Nylon 44 (67.7%) 118 1 by multiple risk factors in male such as cigarette smok-
(63.8%) ing and HIV. Previous studies have shown that patients
Silk 10 (15.4%) 9 (4.9%) 0.16 2.08 0.73-5.94 with pre-morbid illnesses, such as diabetes mellitus are
Vicryl 11 (16.9%) 57 (30.8%) 0.01 0.33 0.13-0.77 at high risk of developing SSI due to their low immunity
Duration of [18]; this was confirmed in this study. As described pre-
operation
viously cigarette smoking was significantly found to be
< 3 hours 43 (66.1%) 163 1
(88.1%) associated with SSI in the multivariate analysis [19].
≥ 3 hours 22 (33.9%) 22 (11.9%) < 0.001 3.8 1.92-7.48 Cigarette smoking has been reported to have an impact
Number of people in
on wound healing through impairment of tissue oxyge-
theatre nation and local hypoxia via vasoconstriction [20].
≤6 11 (16.9%) 137 (74%) 1 The prevalence of HIV in this study was found to be
≥7 54 (83.1%) 48 (26%) < 0.001 14 6.77-29 higher than in general population this has been observed
GA: General anesthesia, SAB: spinal anesthesia block, OR: Odd ratio, CI:
previously [21]. In the present study, the rate of SSI was
confidence interval found to be significantly higher in HIV positive patients
than non HIV patients (p-value < 0.001). Also higher
specimens from different patients had positive bacterial rate of SSI was observed among HIV patients with CD
growth within 48 hours of incubation. Only five out of count below 200 cells/μl (p-value = 0.0001).
56 cultured specimens (8.9%) had mixed growth. Com- Prolonged pre-operative duration of hospitalization
mon bacteria isolated after various procedures (table 1) with exposure to hospital environment has been
were: Staphylococcus aureus 16 (28.6%), Escherichia coli reported to increase the rate of surgical infection [22].
14 (25.0%) and Klebsiella pneumoniae 10 (17.9%), while In this study a hospitalization of more than 7 days prior
the least isolated bacteria was Acinetobacter spp 1(1.8%). to surgery increased the risk of SSI by 2 fold. However
Nine (65%) and 8 (80%) of Escherichia coli and Kleb- in the present study, there were no significant differ-
siella pneumoniae were found to be Extended Spectrum ences in the rate of SSI between patients who received
Beta -Lactamases (ESBL) producer respectively i.e resis- antibiotic prophylaxis and those who didn’t. Despite
tant to first, second, third and fourth generation cepha- lacking of significant association between preoperative
losporins. Most of these ESBL producing isolates were antibiotics and SSI in this study, the authors still believe
isolated from the patients with laparotomy and open that antibiotic prophylaxis is most effective in
Mawalla et al. BMC Surgery 2011, 11:21 Page 6 of 7
http://www.biomedcentral.com/1471-2482/11/21

preventing surgical site infections when administered 30 common isolate in patients who underwent excisional
to 60 minutes before the start of surgery. The lack of biopsy, thoracotomy, mastectomy and appendicectomy.
significance could partly be explained by a non existing Escherichia coli and Klebsiella pneumoniae were the
antibiotic policy regarding different procedures in these most common gram-negative bacteria and were predo-
patients; only 16.4% of patients received pre-operative minantly isolated from in laparotomy and open prosta-
antibiotic prophylaxis. The findings of this study neces- tectomy; similar findings have been reported in other
sitate the introduction of evidence based antibiotic pol- studies [32,33]. The study also found that most of the
icy in this hospital and other hospitals in developing pathogens were multiply resistant to the commonly pre-
countries. scribed antibiotics such as ampicillin, amoxycillin-clavu-
In this study in the univariate analysis it was observed lanic acid, co-trimoxazole, tetracycline, penicillin,
that, the rate of SSI was significantly associated with gentamicin, erythromycin, and ceftriaxone. Other
ASA classification; similar observations have been made authors in Nigeria and Kenya reported a similar antimi-
by other studies [23-25]. In this study ASA class 3 crobial susceptibility pattern [34,35]. These findings
increased the risk for SSIs 45 times. reflect the widespread and indiscriminate use of antibio-
The number of people in theatre during operation has tics, coupled with poor patient compliance and self-
been considered as an independent predictor of SSI treatment without prescription among African patients.
[26]. In the present study, the number of people more The majority of gram negative isolates were sensitive to
than 7 in theatre was significantly associated with meropenem while gram positive being sensitive to van-
increased risk of SSI by 14 times. Also the study comycin and clindamycin; this could be explained by
observed that the use of povodine alone was found to the fact that these antibiotics are relatively rare in the
be associated with higher SSI rates in both univariate hospital and are more expensive so they are rarely
and multivariate analysis than when used in combina- misused.
tion with either chlorhexidine or alcohol related solu-
tions. Povodine iodine has a shorter activity than Conclusion
chlorhexidine and it is inactivated by blood and serum Surgical site infections are a major problem in the surgi-
protein [27,28]. In addition the use of surgical drain has cal wards at Bugando Medical Centre and its incidence
been reported to be associated with an increased risk of is higher than that reported in developed countries.
SSI which was confirmed in this study [24,27,28]. Multi-resistant Staphylococcus aureus followed by
In agreement with other studies [29], the present Escherichia coli and Klebsiella pneumoniae are common
study found that a length of operation of more than 3 bacteria causing SSIs at BMC. Pre-morbidity, use of
hours leads to 4 times higher risk for SSI. Increasing the drain, use of iodine alone in skin preparation, duration
length of procedure theoretically increases the suscept- of operation of more or equal 3 hours and cigarette
ibility of the wound by increasing bacterial exposure and smoking were significantly found to predict SSI at BMC
the extent of tissue trauma (more extensive surgical pro- on multivariate analysis. A better surveillance system for
cedure) and decreasing the tissue level of the prophylac- SSI with feedback of appropriate data to surgeons is
tic antibiotic. highly recommended to reduce the SSI rate in develop-
Surgical wound classification has long been established ing countries.
as an important predictor of the postoperative surgical
site infections [12,30,31]. In our study as in previous
Acknowledgements
studies the risk of SSI was statistically higher in con- The authors would like to acknowledge the technical support provided by
taminated wounds than in clean and clean contaminated the members of the Departments of Microbiology/Immunology and Surgery.
wounds. We thank Mary Louise Shushu, Hezron Bassu and Alpha Boniface for their
excellent technical assistance. This work was supported from research grant
Staphylococcus aureus was the commonest isolate for of WBUCHS to SEM and WM.
the postoperative wound infections. This is consistent
with reports from other studies [5,32]. It has been found Author details
1
Department of Surgery Weill Bugando University College of Health
that in clean surgical procedures, Staphylococcus aureus Sciences, Mwanza, Tanzania. 2Department of Microbiology/Immunology Weill
from the exogenous environment or the patient’s skin Bugando University College of Health Sciences, Mwanza, Tanzania. 3Institute
flora is the usual pathogen, whereas, in other categories of Medical Microbiology, Giessen, Germany.
of surgical procedures, including clean-contaminated, Authors’ contributions
contaminated and dirty, the polymicrobial flora closely BM participated in collecting specimens, collecting clinical data and follow
resembling the normal endogenous microflora of the up of the patients, SEM participated in the design and execution of the
work, performed, microbiological procedures, data analysis, interpretation of
affected site is the most frequently isolated pathogens data and preparation of the manuscript, PLC participated in collecting
[32]. Here, Staphylococcus aureus was the most clinical data and manuscript writing, CI interpretation of the data and
Mawalla et al. BMC Surgery 2011, 11:21 Page 7 of 7
http://www.biomedcentral.com/1471-2482/11/21

manuscript writing and WM designed the study. All authors have read and 23. Haynes SR, Lawler PG: An assessment of the consistency of ASA physical
approved the final manuscript. status classification allocation. Anaesthesia 1995, 50:195-9.
24. Kaye KS, Sands K, Donahue JG, Chan KA, Fishman P, Platt R: Preoperative
Competing interests Drug Dispensing as Predictor of Surgical Site Infections. Emerg Infect Dis
The authors declare that they have no competing interests. 2001, 7:57-65.
25. Haley RW, Culver DH, Morgan WM, White JW, Emori TG, Hooton TM:
Received: 23 December 2010 Accepted: 31 August 2011 Identifying patients at high risk of surgical wound infection. A simple
Published: 31 August 2011 multivariate index of patient susceptibility and wound contamination.
Am J Epidemiol 1985, 2:206-15.
References 26. Kalmeijer MD, van Nieuwland-Bollen E, Bogaers-Hofman D, de Baere GA:
1. Martone WJ, Nicholas RL: Recognition, prevention, Surveillance and Nasal carriage of Staphylococcus aureus is a major risk factor for surgical
Management of SSI. Clin Infect Dis 2001, 33:67-8. -site infections in orthopedic surgery. Infect Control Hosp Epidemiol 2000,
2. Yalcin AN, Bakir Mi Bakici Z, Dokmetas I, Sabir N: Postoperative wound 2:319-23.
infection. J Hosp Infect 1995, 29:305-9. 27. Byrne DJ, Phillips G, Napier A, Cuschieri A: The effect of whole body
3. Brown S, Kurtsikahvi G, Alonso EJ, Aha L, Bochoidez T, Shushtakashiri M, disinfection on intraoperative wound contamination. J Hosp Infect 1991,
Imnadre P: Prevalence and predictors of SSI in Tbilisi Republic of 18(2):145-8.
Georgia. J Hosp Infect 2007, 66:160-166. 28. Kaul AF, Jewett JF: Agents and techniques for disinfections of skin. Surg
4. Berard F, Gordon J: Postoperative wound infections. The influence of Gynecol Obstet 1981, 152:677-685.
ultraviolet infections of the operating room and of various other factors. 29. Kaiser AB, Herrington JL, Jacobs JK: Cefoxitin versus erythromycin,
Ann Surg 1964, 160:1-132. neomycin and cefazolin in colorectal operation. Importance of duration
5. Agarwal SL: Study of postoperative wound infection. Indian J Surg 1972, of surgical procedure. Ann Surg 1983, 198:; 525-530.
34:314-20. 30. Garibaldi RA, Cushing D, Lever T: Risk factors for post operative infection.
6. Rao AS, Harsha M: Postoperative wound infection. J Indian Med Assoc AM J Med 1991, 91:158-163.
1975, 64:90-3. 31. Richard PE: Surgical site infection prevention and control. An emerging
7. Cruse Peter JE, Ford R: The epidemiology of wound infection. A 10 year Paradigm. J Bone Joint Surg Am 2009, 91:2-9.
prospective Study of 62,939 wounds. Surg Clin North Am 1980, 60:27-40. 32. Nichols RL: Surgical Wound Infection. Am J Med 1991, 91:545-635.
8. Tripathy BS, Roy N: Post operative wound sepsis. Indian J Surg 1984, 33. Kotisso B, Assefa A: Surgical Wound Infections in a Teaching Hospital in
47:285-8. Ethiopia. East Afr Med J 1998, 75(7):402-5.
9. Olson MM, Lee JT: Continuous, 10 year wound infection surveillance. 34. Omari MA, Malonzo IM, Bwayo JJ, Mutere AN, Murage EM, Mwatha AK,
Result advantages and unanswered questions. Arch Surg 1990, Ndiya-Achola JO: Pattern of Bacterial Infection and Antimicrobial
125:794-803. Susceptibility at Kenyatta National Hospital, Nairobi, Kenya. East Afr Med
10. Gaynes RP, Culvar TC, Edwards SR, Richards C, Telson JS: Surgical site J 1997, 74(3):134-137.
infection [SSI], rate in the United States 1992-1998. The National 35. Urassa WK, Kagoma G, Lungeland N: Antimicrobial Susceptibilty of
Nosocomial Surveillance System Basic SSI risk index. Clin Infect Dis 2007, Staphylococcus Aureus Strain at Muhimbili Medical Centre, Tanzania.
33:69-77. East Afr Med J 1999, 76(12):693-5.
11. Jan F: Risk factors for SSI in Tanzania district Hospital. A challenge for
the tradition National Nosocomial infection surveillance system index. Pre-publication history
Infect Control Hosp Epidemiol 2006, 27:1401-1404. The pre-publication history for this paper can be accessed here:
12. Erikson HMI, Chugulu S, Kondo Lingaas E: Surgical site infection at KCMC. http://www.biomedcentral.com/1471-2482/11/21/prepub
J Hosp Infect 2003, 55:14-20.
doi:10.1186/1471-2482-11-21
13. Lyamuya EF, Aboud S, Urassa WK, Sufi J, Mbwana J, Ndugulile F,
Cite this article as: Mawalla et al.: Predictors of surgical site infections
Massambu C: Evaluation of simple rapid HIV assays and development of among patients undergoing major surgery at Bugando Medical Centre
national rapid HIV test algorithms in Dar es Salaam, Tanzania. BMC Infect in Northwestern Tanzania. BMC Surgery 2011 11:21.
Dis 2009, 9:19.
14. Mshana SE, Kamugisha E, Mirambo M, Rambau P, Chalya P, Mahalu W,
Lyamuya EF: Prevalence of Clindamycin inducible resistance among
Methicillin-resistant Staphylococcus aureus at Bugando Medical Centre
(BMC), Tanzania. TJ HR 2009, 11:59-64.
15. Mshana SE, Kamugisha E, Mirambo M, Chakraborty T, Lyamuya EF:
Prevalence of multiresistant Gram-negative organisms in a tertiary
hospital in Mwanza, Tanzania. BMC Research Notes 2009, 2:49.
16. Clinical and Laboratory Standards Institute: Performance standards for
antimicrobial disk susceptibility tests. Approved standard. Ninth edition
Document M2-A9 Clinical and Laboratory Standards Institute, Wayne, PA;
2006.
17. Haley RW, Hooton TM, Culver DH: Nosocomial infections in US hospitals,
1975-1976: Estimated frequency by selected characteristics of patients.
Am J Med 1981, 70:947-59.
18. Delamaire M, Maugendre D, Moreno M, Le Goff MC, Allannic H, Genetet B: Submit your next manuscript to BioMed Central
Impaired leucocyte functions in diabetic patients. Diabet Med 1997, and take full advantage of:
14:29-34.
19. Nagachinta T, Stephens M, Reitz B, Polk BF: Risk factors for surgical wound • Convenient online submission
infection. Missouri cardiac surgery. J Infect Dis 1987, 156:967-73.
20. Jone SK, Tripleff RG: The relationship of cigarette smoking to impaired • Thorough peer review
intra-oral wound healing: a review evidence and implication for patient • No space constraints or color figure charges
care. J oral Maxillo Surg 1992, 50:237-40.
• Immediate publication on acceptance
21. Pan A: Surgical site infection in HIV -infected patients from the master
cohort. Clin Infect Dis 2004, 33(2):624-26. • Inclusion in PubMed, CAS, Scopus and Google Scholar
22. Altemeier WA, Calbertson WR, Hummel RP: Surgical consideration of • Research which is freely available for redistribution
endogenous infection- sources, types a method of control. Surg Clin
North Am 1968, 48:227.
Submit your manuscript at
www.biomedcentral.com/submit