A Randomized
Clinical Trial Comparing Partial and Complete Wound Closure in the Prevention
of Surgical Site Infection among Adult Patients Undergoing
Emergency
Colon Resection
Alfred D.
Troncales, MD
Ma. Cecilia T. Leyson, MD
Rodney L.
Dofitas, MD, FPCS
Reynaldo O.
Joson, MD, MHA, MHPEd, MS Surg
Abstract:
Background: In the Department of
Surgery, Hospital X, the incidence of surgical site infection after emergency
colon resection has been a primary concern.
Despite the use of antibiotics and other prophylactic maneuvers, it
still belongs to the top 3 causes of post-operative morbidity as of 2001.
Objective: To determine if the type of wound
closure could affect the incidence of surgical site infection (SSI) in
emergency colon resection we compared the incidence of SSI between partial
(PWC) and complete wound closure (CWC) and analyzed potential factors that may
have contributed to any difference.
Methods: A randomized clinical
trial was conducted from January 2002 to June 2003 wherein patients undergoing
emergency colon resection were closed either partially at the skin-subcutaneous
layer or completely and the wound observed for 28 days. Purulent discharge from the incision site
was used as the indicator of SSI.
Results: A total of 29 patients was
accrued, 14 of whom had CWC while 15 had PWC.
The incidence of SSI was 57 per cent and 20 per cent in CWC and PWC
respectively. Among the factors
analyzed, there was a tendency for physical status, degree of obstruction, and
left-sided colon resection to be contributors to the high SSI rates in CWC.
Conclusion: For patients
undergoing emergency colon resection, partial wound closure seems to be an
effective method in minimizing the risk of
developing surgical site
infection. A multi-center study will be
undertaken to accrue more patients to validate the results of this study.
Key Words: Surgical site
infection, partial wound closure, complete wound closure, emergency colon
resection
Introduction
The
incidence of infection of abdominal incisions after emergency colon resection
was reported to be in the range of 20 to 40 per cent1 despite the
use of antibiotics and other prophylactic maneuvers.
In
the institution of the authors, surgical site infection ranked first as a cause
of post-operative morbidity for the past 2 years (2000 and 2001) and emergency
colon surgery was among the top 3 procedures developing surgical site
infection. Infection developed despite
the use of antibiotics and other prophylactic maneuvers like copious washing
and protection of wound from contamination.
During
the monthly and annual audit conferences, the staff poised the question - are
there other maneuvers that can be done to avoid the development of surgical
site infection in emergency colon resections.
The usual practice had been to close all wounds primarily, that is, a
complete skin-subcutaneous closure after fascial suturing. After brainstorming and consensus gathering,
a trial comparing partial and complete wound closure was decided upon. The main reasons for considering a partial
wound closure were: 1) leaving skin-subcutaneous wounds open is the best
maneuver to avoid a surgical site infection in dirty operative procedures; 2) a
fully gaping skin-subcutaneous wound is an objectionable sight from the point
of view of patients; and 3) a partially gaping skin-subcutaneous wound may be
easier to accept. Thus, the
primary objective of the study was to compare the prevalence rates
of surgical site infection in partial and complete abdominal skin-subcutaneous
wound closure after emergency colon resections. A secondary objective is to analyze for potential factors that
might contribute to any difference that may occur.
Methods
The
study was a randomized clinical trial conducted from January 2002 to June 2003
which included all patients more than 18 years old who underwent emergency
colon resection. Emergency meant
operation done within 24 hours after diagnosis of a surgical disorder.
Before
the implementation of the study, patients were already randomized into partial
and complete wound closure using a “drawing-lot” method. In consideration of the scarcity of
prospective subjects for the study in the Department, an initial trial sample
size of 15 patients per arm was decided upon. Partial wound closure (PWC) was
defined as closure of the skin and subcutaneous tissue in one layer with a 2-0
non-absorbable monofilament suture interruptedly placed 2 inches apart. Complete wound closure (CWC) on the other
hand was closure of skin and subcutaneous tissue with a 4-0 non-absorbable
monofilament suture interruptedly placed 2 centimeters apart. All patients received pre-operative
antibiotics of metronidazole 500 mg intravenously (IV) and gentamycin 80 mg IV
within one hour prior to anesthestic induction. These antibiotics were continued up to the third postoperative
day and then discontinued. Patients
were evaluated daily for one week while still confined in the hospital and then
weekly for four weeks. Surgical site
infection was determined by the presence of purulent drainage, without
laboratory confirmation2, from the incision site. There was no surgical site infection if
there was no purulent discharge from the incision site up to the 28th
day post-operative.
The
operations were done by different senior residents. Follow-up for SSI was done by the junior authors. Due to the small sample size, differences
and trends were determined in terms of prevalence rates and other descriptive
statistics and not analytical statistics.
Results
A
total of 29 patients who underwent emergency colon resection were accrued as of
June, 2003. Of this, male patients
comprised majority of the subjects (25 males vs. 4 females). Fourteen of the 29 patients underwent
complete wound closure (CWC) while the remaining 15 had partial wound closure
(PWC) .
The subjects
in the two arms were more or less comparable in terms of sex and age
distribution (Table 1); co-morbidities (Table 2); physical status (Table 3);
side of colon resection (Table 4); and presence and degree of intestinal
obstruction (Table 5).
The
SSI rate in CWC was 57 per cent while that in PWC was 20 per cent (Table
6). There was no mortality and no
anastomotic leak in these 29 patients.
Resection-anastomosis was done in 83 per cent of cases.
Attempts
were done to analyze for factors that might contribute to the high SSI rates
specifically physical status, presence of co-morbid conditions, side of colon
resection, and presence and degree of intestinal obstruction.
Physical
status (PS):
The
overall SSI rates were 100 per cent for PS III-E, 50 per cent for PS II-E and
20 per cent for PS I-E (Table 7). There
was a tendency for SSI rates to rise with increasing PS, overall as well as in
the two arms (CWC and PWC). There was
also a tendency for PWC as compared to CWC to reduce the SSI rates in PS I-E
and PS II-E.
Co-morbid
conditions:
The
overall SSI rates were 75 per cent in those patients with co-morbid conditions
and 24 per cent in those without (Table 8).
There was a tendency for SSI rates to rise in the presence of co-morbid
conditions. There was a tendency for
PWC to reduce the SSI rates in patients with no co-morbid conditions but not in
those with co-morbid conditions.
Side of colon
resected:
For
ease of analysis, the transverse and left-sided resections were lumped together
as left-sided resections. The overall
SSI rates were 45 per cent and 33 per cent for left-sided and right-sided colon
resections respectively (Table 9).
There was a tendency for PWC to reduce the SSI rates in both left and
right colon resection.
Presence of
intestinal obstruction:
The
overall SSI rates were 48 per cent, 25 per cent, and 17 per cent in those with
complete, partial, and without intestinal obstruction respectively (Table
10). There was a tendency for PWC to
reduce the SSI rates in patients with complete intestinal obstruction.
Discussion
Surgical
site infection (SSI) after emergency colon resection is a concern among
surgeons and more so, the patients.
Firstly, by itself, SSI is an unwanted adverse event. Secondly, the reported incidence could reach
as high as 40 per cent1.
In
the Philippines and even internationally, data on the incidence of SSI after
colon resection, both elective and emergency, are scarce, and if present,
variable, more so on the emergency cases.
In view of the absence of reliable data, the authors in analyzing and
discussing the results of this study would just have to contend with whatever
data are available in the literature and data gotten from studies on elective
colon surgery or combined elective and emergency cases, with the assumption
that the incidence of SSI is higher in emergency than in elective cases.
In
a 1994 report by the Philippine College of Surgeons (PCS) on postoperative
infection rates in six pilot hospitals4, the overall SSI rate after
colorectal surgery was 7.5 per cent, with a range of 2.2 to 13.8 per cent from
one hospital to another. These figures
covered both elective and emergency operations and all sorts of colorectal
operations including those for hemorrhoids and fistulas-in-ano, making the
figures on the low side if only colorectal resections are considered, and more
so, if limited to emergency cases.
Internationally,
the data that the authors decided to use for this analysis and discussion came
from two meta-analysis papers on the benefit of antibiotics
in colorectal surgery. Baum in 19815, utilizing 26
trials from 1965 to 1980, reported a 36 per cent infection rate in the control
group versus 22 per cent in the group with antibiotics. Song in 19986 utilizing 147
randomized clinical trials from 1984 to 1995 reported a 40 per cent wound
infection rate in the control group versus 13 per cent in the antibiotic
group. Thus, the incidence of SSI after
colorectal surgery could really reach as high as 40 per cent. This figure could be higher in emergency
colon resections.
In
Hospital X Surgery, since 2001, controlling wound infection or SSI has been
included in its wide-scale quality improvement program. It started with a significant reduction of
SSI in ruptured appendicitis by leaving the skin-subcutaneous layer
unsutured. From this experience, the
approach was considered for another condition with significantly high SSI rate,
that of emergency colon surgery. Thus,
this study was done to test a strategy to reduce the incidence of SSI in
emergency colon resection. Leaving the
skin-subcutaneous layer is the best guarantee against SSI; however, this might
not be readily acceptable to the patients.
Thus, a trial of partial wound closure.
At
this point of the trial, although the sample size is still small, the results
tended to show that PWC significantly reduces SSI after an emergency colon
resection and therefore, could be a useful strategy. In support of this recommendation for using PWC in reducing SSI
are the following findings and arguments:
1. In a study
comparing laparoscopy (LCR) and open colonic resection (OCR), SSI occurred in
13.5 per cent of patients after LCR and in 10.9 per cent of patients after OCR7. The extraction site for LCR was associated
with a high incidence of complications compared to OCR. The explanation was that the incision in the
extraction site in LCR was more heavily contaminated and more air-tight in
closure as compared to the bigger incisional wound in OCR.
2. The almost 50
per cent reduction of SSI rates in PWC (20 per cent vs. 57 per cent in CWC) in
this study parallels the same degree of reduction of SSI rates in those given
antibiotics as shown in the meta-analysis of Baum (5) and Song (6). This
suggests that PWC is a very strong factor and strategy in SSI reduction,
approximating the efficacy of leaving the skin-subcutaneous layer unsutured in
ruptured appendicitis and heavily contaminated abdominal wounds.
3. Trying other
more efficacious but more expensive antibiotics against gram-negative and
anaerobic organisms might be an option that the authors considered. However, considering the financial
constraints of the setting (government hospital as well as patients in the
lower income strata), this option is not a sustainable proposition.
4. Mechanical
bowel preparation even if feasible in emergency cases, in a 2003 meta-analysis
of 6 randomized control trials (8) did not show significant reduction in wound
infection rate (7.4 per cent vs. 5.7 per cent in those without preparation).
5. There are no
randomized control trials on intra-operative colonic irrigation in the
prevention of wound infections. The
reported wound infection rates associated with the use of intra-operative
colonic irrigation are high at 10.81 per cent (9) and 29 per cent (10). As to the risk factors for surgical site
infection, in a study by Tang (11) in 2,809 consecutive patients undergoing
elective colorectal resection via laparotomy between February 1995 and December
1998, the following were identified as overall risk factors: American Society
of Anesthesiology (ASA) or Physical Status score 2 or
3, male
gender, surgeons, types of operation, creation of an ostomy, contaminated
wound, use of drainage, and intra- or postoperative blood transfusion. In the study of Mandal (12) in colonic
trauma, risk factors identified were left-sided colon injuries and ineffective
antibiotics against anaerobic bacteria.
Statistical analysis on the different surgical methods used such as
primary repair, primary repair with exteriorization, or colostomy, was
indeterminable. In this study, because
of the small sample size, no definite conclusion can be made as to the risk
factors in SSI in emergency colon resections. However, the more reliable potential risk factors noted in this
study were the physical status, degree of intestinal obstruction and left sided
colon resection which may implicate degree of contamination of the surgical
wound during the resection. In conclusion,
this pilot study shows that for patients undergoing emergency colon resection,
partial wound closure seems to be an effective method in minimizing the risk of
developing surgical site
infection. A multi-center study will be undertaken to
accrue more patients to validate the results of this study.
References:
1. Dunn DL.
Surgery: Basic Science and Clinical Evidence, 1st ed. United States:
Springer Verlag New York, Inc., 2001: 205-208
2. Mangram AJ,
Horan TC, Pearson ML, Silver LC, Jarvis WR. Guidelines for the Prevention of
Surgical Site Infection, 1999. Am J of
Infection Control 1999: 27: 97-134
3. Howard RJ.
Principles of Surgery, 7th ed. Singapore: McGraw-Hill Book Co, 1999:
130-131.
4. Cortez ER;
Ampil IDE; Laudico AV et al. Postoperative infection
rates: a report on the PCS surgical infection surveillance system in six pilot
hospitals. Philipp J Surg Spec 1994; 49(1):37-42.
5. Baum ML, Anon DS, Chalmus TC, Sachs HS, Smith H. A survey of clinical trials of antibiotic prophylaxis in colon surgery: evidence against further use of no treatment controls. N Engl J of Med 1981; 305(14): 795-799.
6. Song F, Glenny
AM. Antimicrobial prophylaxis in
colorectal surgery: a systematic review of randomized control trials. Br J Surg 1998; 85:1232-1241.
7. Winslow
ER, Fleshman JW, Birnbaum EH, Brunt LM. Wound complications of laparoscopic vs
open colectomy. Surg Endosc 2002;16(10):1420-5.
Guenaga KF, Matos D, Castro AA, Atallah AN, Wille-Jorgensen P. Mechanical bowel preparation for elective colorectal surgery. Cochrane Database Syst Rev 2003;(2):CD001544.
9. Sule A, Obepka PO, Iya D, Ogbonna B, Momoh J. Intraoperative colonic
irrigation in the management of left sided large bowel emergencies in Jos
University
Teaching Hospital, Nigeria. East Afr Med J 2000;77(11):613-7.
10. Biondo S, Perea MT, Rague JM, Pares D, Jaurrieta
E. One-stage procedure
in non-elective surgery for diverticular disease complications. Colorectal Dis
2001;3(1):42-5.
11. Tang R, Chen HH, Wang YL et al. Risk
factors for surgical site infection after elective resection of the colon and
rectum: a single-center prospective study of 2,809 consecutive patients. Ann
Surg 2001;234(2):181-9.
12. Mandal AK, Thadepalli
H, Matory E, Lou MA, O'Donnell VA Jr.
Evaluation of
antibiotic therapy and surgical techniques in cases of
homicidal wounds of
the colon. Am Surg 1984;50(5):254-7.
Table 1. Sex and age distribution of
patients under the two arms.
|
CWC N= 14 |
PWC N=15 |
M:F |
11:3 |
14:1 |
Age range |
18-72 y/o |
24-61 y/o |
Mean age |
37 y/o |
38 y/o |
> 45 years
old |
5 |
5 |
< 44
years old |
9 |
10 |
Table 2. Co-morbid condition.
Co-Morbidities |
CWC |
PWC |
1. None |
10 |
11 |
2. Presence Hypertension Pulmonary TB |
4 2 2 |
4 4 0 |
Total |
14 |
15 |
Table 3.
Physical status.
Physical
Status |
CWC |
PWC |
I-E |
7 |
8 |
II-E |
7 |
5 |
III-E |
- |
2 |
IV-E |
- |
- |
Total |
14 |
15 |
Table 4. Side of colon resected.
Side of
colon resected |
CWC |
PWC |
1. Right Colon |
7 |
11 |
2. Left Colon |
5 |
- |
3. Left and Transverse Colon |
- |
2 |
4. Transverse |
2 |
2 |
Total |
14 |
15 |
Table 5. Degree of obstruction.
Degree of
Obstruction |
CWC |
PWC |
1. Complete Obstruction |
10 |
9 |
2. Partial Obstruction |
2 |
2 |
3. Non-obstructed |
2 |
4 |
Total |
14 |
15 |
Table 6. Surgical site infection
rate.
Wound closure |
Total Number |
Infection |
Infection
Rate |
1. Complete |
14 |
8 |
57% |
2. Partial |
15 |
3 |
20% |
Total |
29 |
11 |
100% |
Table 7. SSI rates based on physical
status.
Physical status |
Overall SSI
rates |
SSI rates in
CWC |
SSI rates in
PWC |
I –E |
3/15 (20%) |
3/7 (43%) |
0/8 (0%) |
II – E |
6/12 (50%) |
5/7 (71%) |
1/5 (20%) |
III – E |
2/2 (100%) |
- |
2/2 (100%) |
IV – E |
- |
- |
- |
Table 8. SSI rates based on presence or
absence of co-morbid conditions.
Co-morbid conditions |
Overall SSI
rates |
SSI rates in
CWC |
SSI rates in
PWC |
Absent |
5/21 (24%) |
5/10 (50%) |
0/11 (0%) |
Present |
6/8 (75%) |
3/4 (75%) |
3/4 (75%) |
Table 9. SSI rates based on side of
colon resected.
Side of colon resected |
Overall SSI
rates |
SSI rates in
CWC |
SSI rates in
PWC |
Right side |
6/18 (33%) |
4/7 (57%) |
2/11 (18%) |
Left side |
5/11 (45%) |
4/7 (57%) |
1/4 (25%) |
Table 10. SSI rates based on degree of
intestinal obstruction.
Degree of obstruction |
Overall SSI
rates |
SSI rates in
CWC |
SSI rates in
PWC |
None |
1/6 (17%) |
1/2 (50%) |
0/4 (0%) |
Partial |
1/4 (25%) |
0/2 (0%) |
1/2 (50%) |
Complete |
9/19 (48%) |
7/10 (70%) |
2/9 (22%) |