Objective: We studied the impact of residual bacteria at the incision site after
disinfection with polyvinylpyrrolidone (PVP or povidone)-iodine-alcohol and
the correlation with postoperative surgical site infections (SSIs).
Background: Chlorhexidine-based preparations are significantly more effective
for catheter insertion care than povidone-iodine solutions to prevent
catheter-associated infections, suggesting that the use of PVP-iodine should
be reevaluated for disinfection of the surgical site. In the majority of European
hospitals PVP-iodine-alcohol is still standard of care to prepare the
preoperative site.
Methods: We consecutively and prospectively enrolled 1005 patients from
representative surgical disciplines. Skin cultures to determine skin microbial
counts were taken after disinfection with PVP-iodine-alcohol, immediately
before incision. Disinfection of the surgical sitewas performed using standardized
procedure under supervision. Criteria for SSI were based on guidelines
issued by the Centers for Disease Control including appropriate follow-up of
30 days and 1 year.
Results: A total of 1014 skin cultures from surgical sites were analyzed from
1005 patients, of which 36 (3.6%) revealed significant colonization of the
preoperative site, and 41 SSIs were detected, accounting for an SSI rate of
4.04%; residual bacteria before incision were completely unrelated to the
incidence of SSI, even after adjustment for multiple potentially confounding
variables.
Conclusions: A low rate of SSIs of 4.04% was achieved when using PVPiodine-
alcohol for disinfection of the preoperative site. Remaining bacteria
after standardized 3-step disinfection did not at all correlate with the development
of an SSI. Our data provide clear evidence that PVP-iodine-alcohol is
effective for preparation of the preoperative site.
(Ann Surg 2012;255:565–569)
Surgical site infections (SSIs) are the most common hospitalacquired
infections in surgical departments. They not only lead
to significant increases in morbidity and mortality,1,2 but also to
significant growth of health care expenditures.3,4 The most common
pathogens involved are Staphylococcus aureus and coagulasenegative
staphylococci,5,6 components of normal skin flora.7 Therefore,
preoperative disinfection of the surgical site with an antiseptic
From the ∗Divisions of Infectious Diseases and Hospital Epidemiology; †Clinical
Microbiology; and ‡Department of Surgery, University Hospital, Basel,
Switzerland. Ms Egly-Gany is currently employed at B. Braun Medical
AGSeesatz 176204 Sempach, Switzerland.
Disclosure: No author has any conflict of interest. The University Hospital of Basel
funded in part the microbiological analyses of the samples as part of the quality
improvement program. No other funding was received for the study.
Reprints: Andreas Franz Widmer, MD, MSc, Division of Infectious Diseases and
Hospital Epidemiology, University Hospital Basel, Petersgraben 4, CH-4031
Basel, Switzerland. E-mail: widmerA@uhbs.ch.
Copyright C 2012 by Lippincott Williams & Wilkins
ISSN: 0003-4932/12/25503-0565
DOI: 10.1097/SLA.0b013e3182468b2d
skin preparation is standard practice before any surgical intervention
to decrease skin microbial counts before incision. To date, the
Centers for Disease Control and Prevention (CDC) and the World
Health Organisation have not released any recommendations as to
which antiseptic should preferably be used to prevent postoperative
SSIs. However, the CDC does recommend 2% chlorhexidine-based
preparations for cleaning the site of insertion of vascular catheters.8
As compared with PVP-iodine, a chlorhexidine–alcohol solution has
been found to reduce catheter-associated infections by approximately
50%.9 Such data suggest that chlorhexidine solutions may also be
more effective than PVP-iodine in preparing the preoperative site.
At our hospital—as in the majority of European hospitals—it
is standard of care to prepare the preoperative site with PVP-iodinealcohol.
Therefore, we initiated a study to evaluate the impact of
residual bacteria at the incision site after disinfection with PVPiodine-
alcohol and the correlation with postoperative SSIs in a large
trial.
METHODS
Setting
The University Hospital of Basel is a tertiary care center of
855 beds in Basel, Switzerland, with approximately 28,000 surgical
interventions on inpatients.
Study Design
The data for this prospective observational study were collected
from November 2008 to April 2010. The study was approved
by the institutional review board as part of the quality improvement
program, supported by the hospital executive board. All inpatient
procedures performed in the divisions of vascular, visceral, cardiac,
thoracic, orthopedic, neurological, and gynecological surgery were
consecutively enrolled. The investigators had no influence on the
participation of patients enrolled in the study. Patients were subsequently
observed for the development of an SSI. Data were analyzed
to examine the association between microbial flora of the skin after
disinfection and the presence of an SSI.
Procedure
Disinfection of the surgical site was performed according to
the study protocol, using standardized procedure under supervision of
a study nurse or the attending surgeons. The applied iodine-alcohol
product contained 50 g of propan-2-ol and 1 g of PVP-iodine per
100 g (Braunoderm; B. Braun Medical AG, Sempach, Switzerland).
Skin Microbial Counts
After disinfection of the preoperative surgical site as per hospital
protocol, skin cultures were taken from the predefined incision
site. A template with an area of 20 cm2 was used to standardize the
skin size area to be cultured as previously described.10 The template
areawas swabbedwith a sterile cotton swab, premoistened with sterile
0.9% sodium chloride in a rotating motion. Each swab was immediately
placed in a vial containing 2 mL of a disinfection neutralizing
solution, to avoid carry over effects of residual disinfectant. This solution
is composed of polysorbate 80 30 g/L, lecithin 3g/L, L-histidin
1 g/L, sodium thiosulfate 5 g/L, tripticase soy broth 30 g/L, and dematerialized
water 1L. Swabs were plated onto 5% sheep blood agar
within 24 hours and incubated for 48 hours at 35◦C (±2◦C). Colonyforming
units (CFU) were enumerated and identified by means of
standard laboratory identification methods. In addition, the neutralizing
solution of each sample was incubated for 5 days at 35◦C (±2◦C)
and if applicable, subsequent growth was identified.
According to the number of CFU detected, samples were classified
into 2 categories: “significant growth” and “nonsignificant
growth.” Samples with 10 or more CFU were defined as “significant
growth.”
Surgical Site Infections
The University Hospital participates in the national SSI
surveillance with postdischarge surveillance. For SSI, definitions of
the CDC/National Nosocomial Infection Surveillance were used.11
Infections were classified as (i) superficial incisional (which involved
only skin and subcutaneous tissue and excluded stitch-related abscesses),
(ii) deep incisional (which involved fascia and muscle), or
(iii) organ–space infections (which involved any organ or space other
than the incised layer of body wall that was opened or manipulated
during surgery). In addition, patient demographics and risk factors
for SSI were prospectively collected using a standardized case report
form. These included gender, age, presence of diabetes, smoking (at
the time of surgery), obesity (defined as a body mass index>25), immunosuppressive
medication (≥0.5 mg/kg prednisone, or equivalent
doses of other steroids and other immunosuppressive agents), type of
surgery, American Society ofAnaesthesiologists (ASA) score,wound
contamination class, time of administration of preoperative antibiotic
prophylaxis, and duration of surgery expressed as under or over-limit,
when compared to a cut-point (T time). The administration of preoperative
antibiotic prophylaxis was classified as being correct, when it
was administered 0 to 60 minutes before cut or if the type of surgery
did not require any prophylaxis according to the CDC guidelines.11
Postdischarge surveillance was completed by telephoning patients
after 30 days postoperatively and after 1 year postoperatively
for patients with implants. If patients were lost to follow-up, the decision
on the presence of an SSI was based on medical chart review.
Full medical chart review for the presence of SSI was performed
by 2 board-certified infectious diseases specialists.When SSIs were
suspected or diagnosed, clinically relevant samples were cultured.
Bactericidal Tests
The bactericidal activity was determined using the standard
method for testing bactericidal activity of chemical disinfectants
according to the German Society for Hygiene and
Microbiology.12 The disinfectant Braunoderm (isopropanolic solution
of polyvinylpyrrolidone-iodine) was tested for activity against
3 isolates of S. aureus, 6 isolates of coagulase-negative staphylococci,
and 1 isolate of Pseudomonas aeruginosa. The inactivating
solution used was composed of polysorbate 80 30 g/L, lecithin 3 g/L,
L-histidin 1 g/L, and sodium thiosulfate 5 g/L.
Aquantitative suspension test with an organic load (0.3% albumin
and 0.3% sheeps erythrocytes)was carried out.After an exposure
time of 30 seconds at 20◦C, the disinfectant was inactivated and the
suspension was cultured for 48 hours at 36◦C. If a reduction factor of
5 in the number of CFU was achieved, the disinfectant was regarded
effective, according to EN 1040.
Statistical Issues
Datawere entered into a database (Excel;Microsoft,Redmond,
WA) using TeleFormautomatized data entry (Cardiff Software, Vista,
CA) and then imported into SSPS (Version 120.0; Chicago, IL).
The main outcome of the study is the development of SSI. The chisquare
and Fisher exact test were used to analyze variables. A logistic
regression model was used for the analyses of the significance of the
wound contamination class andASAscore regarding the development
of an SSI. Logistic regression analysis of all variables detected to be
significant in univariate analysis was performed. P values less than
0.05 were regarded as statistically significant.
RESULTS
A total of 1014 skin cultures from surgical sites were analyzed
from 1005 patients. Nine patients had 2 skin cultures taken, but the
samples came from 2 different surgical sites of 2 surgeries during the
same procedure. None of the patients were excluded. The patient’s
median age was 60 years, and there was a slight male predominance
(553, 54.5%) (Table 1). The majority of cases underwent orthopedic
or cardiac surgeries (289, 28.5% and 229, 22.6%, respectively)
(Table 2).
Postdischarge surveillance was completed in 813 (80.12%)
cases by telephoning patients after 30 days postoperatively. Additional
postdischarge surveillance 1 year after the operation was performed
for all patients receiving an implant (465 cases). This could
be completed for 338 (72.67%). Twenty-two patients had died in the
meantime, the cause of death not being an SSI.
Significant bacterial growth after standardized and supervised
disinfection of the preoperative site with PVP-iodine-alcohol was
found in 36 cases (3.6%). Significant growth of bacteria after disinfection
and before incision was not a significant risk factor for the
development of an SSI (P = 0.639) (Table 3). Only 2 cases (2/36;
5.5%) with significant bacterial growth after disinfection developed
an SSI compared with 39 cases (39/978; 3.99%) with SSIs without
significant growth after disinfection. There was an unexpected inverse
relationship between residual bacteria before incision and the
incidence of SSIs, indicating that residual bacteria do not play a significant
role after appropriate disinfection with PVP iodine-alcohol.
Residual bacteria remained not significantly associated with SSIs
even after adjustment of all potentially important confounders (P=
0.645) (Table 4).
The pathogens isolated were coagulase-negative staphylococci
in most cases (30, 83.3%), followed by S. aureus in 3
cases (8.3%), Corynebacterium spp. in 2 cases (5.6%), and Pseudomonas
aeruginosa in 1 case (2.8%). We further analyzed risk factors
for having significant bacterial growth after disinfection of the
TABLE 1. Demographic Data and Baseline
Characteristics of 1014 Cases
Characteristic No. (%)
Age, median (range), y 60 (15–100)
Gender
Male 553 (54.5)
Female 461 (45.5)
Diabetic 156 (15.4)
Smokers 244 (24.1)
Obesity (BMI > 25) 478 (47.1)
Immunosuppressive therapy 53 (5.2)
Admitted from
Home 901 (88.9)
Other hospital 105 (10.4)
Long-term-care facility 8 (0.8)TABLE 2. Perioperative and Postoperative Characteristics
of 1014 Cases
Characteristic No. (%)
Type of surgery
Cardiac 229 (22.6)
Thoracic 73 (7.2)
Vascular 91 (9.0)
Orthopedic 289 (28.5)
Neurological (cranium) 74 (7.3)
Neurological (other than cranium) 99 (9.8)
Gastrointestinal 38 (3.7)
Gynecological 50 (4.9)
Other 71 (7.0)
Emergency surgery 55 (5.4)
Implant 465 (45.9)
Wound contamination class
Class I/clean 926 (91.3)
Class II/clean-contaminated 50 (4.9)
Class III/contaminated 15 (1.5)
Class IV/dirty-infected 23 (2.3)
ASA score
1 93 (9.2)
2 381 (37.6)
3 493 (48.6)
4 47 (4.6)
T time exceeded 226 (22.3)
Reoperation 115 (11.3)
Correct administration of antibiotic prophylaxis
(0–60 minutes before cut)
781 (77.0)
SSI 41 (4.0)
preoperative site with PVP-iodine-alcohol and found a BMI >25
kg/m2 (P=0.013) to be significantly associated with bacterial growth
after disinfection. No bacterial growth was detected in the neutralizing
solution, serving as negative controls (1014/1014).
The rate of SSI during the study period was 4.04% (41/1014).
The majority of SSIs (23, 56.1%) were organ-space infections, 13
(31.7%) were deep incisional, and 5 (12.2%) were superficial SSIs.
Male sex, diabetes, exceeded T time, and reoperations were
significant risk factors for the development of SSI. Increasing wound
contamination class and ASA score were also significantly associated
with the risk for the development of an SSI (Table 3). The majority
of patients (34, 94.4%) with significant growth after disinfection of
the incisional site did not develop an SSI (Table 3).
Significant risk factors for SSI by multivariate analyses were
diabetes, ASA score, and reoperation. There were 2 cases with significant
bacterial growth after disinfection and development of an
SSI: the first developed an organ-space infection of the bone after
an orthopedic intervention with detection of S aureus as causative
pathogen and revealed coagulase-negative staphylococci in the preoperative
culture of the surgical site after disinfection, the second had
a superficial SSI which was treated without a bacterial culture being
performed, the preoperative culture also detected coagulase-negative
staphylococci.
Bactericidal activity of PVP-iodine-alcohol was determined
for all 3 isolates of S aureus, 6 isolates of coagulase-negative staphylococci,
chosen at random from the 30 isolates detected, and for the 1
isolate of Pseudomonas aeruginosa. A reduction factor greater than
5 demonstrating effectiveness was detected for all examined isolates.
DISCUSSION
The overall rate of SSIs was 4.04% using PVP iodine-alcohol,
similar or lower than reported in studies on SSIs with similar procedures.
Only 3.6% or 36 of a total of 1014 cases had significant
bacterial growth after disinfection, and these cases did not have a
higher risk for the development of an SSI as compared with patients
with no detection of significant numbers of CFU after preparation
of the surgical site. This finding is consistent with the results of
Cronquist et al,13 who also found no association of skin microbial
counts after disinfection and a subsequent SSI in 609 neurosurgical
patients. However, another study conducted with 188 cholecystectomy
surgical patients found that bacteria at the incision site make a
substantial contribution to wound flora.14 Because the patient’s skin
is a major source of pathogens responsible for the development of
SSIs, it seems plausible that improving skin antisepsis would decrease
the development of SSIs.15 Our data, however, suggest that
standardized 3-step disinfection with PVP-iodine-alcohol is sufficient
to prepare the surgical site and that remaining bacteria after
this procedure do not constitute a risk factor for the development of
SSIs.
Darouiche et al16 conducted a randomized controlled clinical
trial with 409 subjects in the chlorhexidine alcohol group and 440
in the PVP-iodine group, comparing preoperative cleansing of the
patient’s skin with chlorhexidine alcohol to PVP-iodine, and found
chlorhexidine alcohol to be superior for preventing SSIs. The SSI rate
was 9.5% in the chlorhexidine alcohol group as compared to 16.1% in
the PVP-iodine group—a difference that was statistically significant.
Interestingly, the rate of the most severe SSIs, organ-space infections,
and sepsis resulting from SSI did not differ significantly between the
2 groups.16
Chlorhexidine alcohol and PVP-iodine both possess broadspectrum
antimicrobial activity.11 Darouiche et al16 concluded that
the superiority of chlorhexidine alcohol was related to its more rapid
action, persistent activity despite exposure to bodily fluids, and its
residual effects.17 However, this study was smaller than the present
trial, the disinfection was not supervised and PVP-iodine was not
applied in alcohol as chlorhexidine was. In addition, there were no
microbiological samples taken prior to incision, obviously, the crucial
step in controlling the efficacy of an antiseptic.
Chlorhexidine-based solutions were more effective than
iodine-containing solutions in reducing bacterial concentration of the
surgical site for vaginal hysterectomy18 and foot-and-ankle surgery.19
The former study compared chlorhexidine with PVP-iodine and found
a significantly lower rate of bacterial contamination 30 minutes after
disinfection in the chlorhexidine group. Interestingly, subsequent
cultures, performed 90 minutes and 150 minutes after initial surgical
scrubs, failed to demonstrate significant differences between
the 2 groups. In both studies, success of the different solutions was
merely measured by determination of residual bacterial growth after
disinfection—the rate of SSIs was not assessed. As our data show,
there is however no correlation between residual bacterial growth and
the development of an SSI, limiting the value of this finding.
When testing 3 different types of surgical-site scrubs—PVPiodine
with an isopropyl alcohol, 2% chlorhexidine and 70% isopropyl
alcohol, and iodine povacrylex in isopropyl alcohol—in a
large study comprising 3209 operations, iodine povacrylex in alcohol
was superior compared to the other 2 scrubs in the prevention of SSIs
in general surgery patients. In subgroup analysis, no difference in
outcome was seen between patients prepared with PVP-iodine with
an isopropyl alcohol and those prepared with iodine povacrylex in
alcohol, but patients in both these groups had significantly lower SSI
rates, compared with rates for patients prepared with 2% chlorhexidine
and 70% isopropyl alcohol. The authors therefore concluded
that iodophor-based compounds may be superior to chlorhexidine in
general surgery patients.20
Several limitations of this study should be mentioned. First,
the study was not randomized, but a strict protocol was followed to
consecutively enroll all eligible patients, avoiding a potential bias inTABLE 3. Risk Factors for Development of SSI
SSI, No. (%) No SSI, No. (%)
Risk Factor n = 41 n = 973 P OR (95% CI)
Gender
Male 30 (73.2) 523 (53.8) 0.014 2.28 (1.15–4.49)
Female 11 (26.8) 450 (46.2)
Admitted from
Home 34 (82.9) 867 (89.1) 0.306
Other hospital 7 (17.1) 98 (10.1)
Long-term-care facility 0 (0.0) 8 (0.8)
Emergency surgery 2 (4.9) 53 (5.4) 0.880 0.89 (0.21–3.79)
Receipt of implant 19 (46.3) 446 (45.8) 0.950 1.00 (0.55–1.91)
Diabetes 14 (34.1) 142 (14.6) 0.001 3.00 (1.55–5.93)
Smoking 9 (22.0) 235 (24.2) 0.750 0.88 (0.41–1.88)
Obesity (BMI > 25) 24 (58.5) 454 (46.7) 0.140 1.61 (0.86–3.05)
Immunosuppressive therapy 1 (2.4) 52 (5.3) 0.720 0.44 (0.06–3.05)
Wound contamination class < 0.005
Class I/clean 29 (70.7) 897 (92.2)
Class II/clean-contaminated 4 (9.8) 46 (4.7)
Class III/contaminated 2 (4.9) 13 (1.3)
Class IV/dirty-infected 6 (14.6) 17 (1.7)
ASA score < 0.005
1 0 (0.0) 93 (9.6)
2 9 (22.0) 372 (38.2)
3 26 (63.4) 467 (48.0)
4 6 (14.6) 41 (4.2)
T time exceeded 15 (36.6) 211 (21.7) 0.025 2.10 (1.1–4.0)
Reoperation 31 (75.6) 84 (8.6) < 0.005 32.80 (15.55–69.25)
Correct administration of antibiotic
prophylaxis (0–60 minutes before cut)
34 (82.9) 747 (76.8) 0.36 1.47 (0.64–3.37)
Significant growth after disinfection
(CFU ≥ 10)
2 (4.9) 34 (3.5) 0.639 1.41 (0.33–6.10)
No significant growth after disinfection
(CFU < 10)
39 (95.1) 939 (96.5)
CI indicates confidence interval.
TABLE 4. Multivariate Analysis of Risk Factors for the Development of SSI
CI 95%
Exp(B) Lower Upper Significance (P)
Diabetes 2.401 1.070 5.384 0.034
ASA score 2.190 1.221 3.930 0.009
Reoperation 31.507 14.673 67.654 0.000
Significant growth after disinfection 1.464 0.290 7.403 0.645
CI indicates confidence interval.
selecting patients. Second, the power of this study is limited by the
sample size. However, the study power exceeds 90% to detect differences
if they would exist. Larger randomized trials with appropriate
endpoints, monitored and supervised application of the disinfectant,
postdischarge surveillance of SSI, stratified by BMI are unlikely to
be financed.
In summary, our data provides strong evidence that PVPiodine-
alcohol is effective for the preparation of the preoperative
site. A low rate of SSIs was achieved when using this disinfectant and
remaining bacteria after standardized 3-step disinfection did not correlate
with the development of an SSI. Furthermore, the test results of
the bactericidal activity of PVP-iodine-alcohol revealed effectiveness
of PVP-iodine-alcohol for all examined isolates. We therefore conclude
that remaining bacteria after standardized 3-step disinfection do
not at all correlate with the development of an SSI. Our data provide
clear evidence that PVP-iodine-alcohol is effective for preparation of
the preoperative site.
ACKNOWLEDGMENTS
The University Hospital of Basel funded in part the microbiological
analyses of the samples as part of the quality improvement
program. No other, especially no commercial funding was received
for the study.
REFERENCES
1. Astagneau P, Rioux C, Br¨ucker G. Morbidity and mortality associated with
surgical site infections: results from the 1997–1999 INCISO surveillance. J
Hosp Infect. 2001;48:267–274.
2. Coello R, Charlett A,Wilson J, et al. Adverse impact of surgical site infections
in English hospitals. J Hosp Infect. 2005;60:93–103.3. Weber WP, Zwahlen M, Reck S, et al. Economic burden of surgical site
infections at a European university hospital. Infect Control Hosp Epidemiol.
2008;29:623–629.
4. Dimick JB, Chen SL, Taheri PA, et al. Hospital costs associated with surgical
complications: a report from the private-sector national surgical quality
program. J Am Coll Surg. 2004;199:531–537.
5. Noble W. Skin as a source for hospital infection. Infect Control. 1986;7:111–
112.
6. Van Ek B, Bakker FP, van Dulken H, et al. Infections after craniotomy: a
retrospective study. J Infect. 1986;12:105–109.
7. Murray PR, Baron EJ, Jorgensen JH, Pfaller MA, Yolken RH, eds. Manual of
Clinical Microbiology. 8th ed. Washington, DC: ASM Press; 2003.
8. O’Grady NP, Alexander M, Dellinger EP, et al. Guidelines for the prevention
of intravascular catheter-related infections. Infect Control Hosp Epidemiol.
2002;23:759–769.
9. Chaiyakunapruk N, Veenstra DL, Lipsky BA, et al. Chlorhexidine compared
with povidone-iodine solution for vascular catheter-site care: a meta-analysis.
Ann Intern Med. 2002;136:792–801.
10. TietzA, FreiR,DangelM, et al. Octenidine hydrochloride for the care of central
venous catheter insertion sites in severely immunocompromised patients. Infect
Control Hosp Epidemiol. 2005;26:703–707.
11. Mangram AJ, Horan TC, Pearson ML, et al. Guideline for prevention of
surgical site infection, 1999. Infect Cotrol Hosp Epidemiol. 1999;20:250–
278.
12. German Society for Hygiene and Microbiology. Pr¨ufung und Bewertung
chemischer Desinfektionsverfahren—Stand 12.07.1991. Hyg Med.
1991;Sonderdruck:1–8.
13. Cronquist AB, Jakob K, LaiL, et al. Relationship between skin microbial counts
and surgical site infection after neurosurgery. Clin Infect Dis. 2001;33:1032–
1038.
14. Whyte W, Hambraeus A, Laurell G, et al. The relative importance of routes
and sources of wound contamination during general surgery. J Hosp Infect.
1991;18:93–107.
15. Napolitano LM. Decolonization of the skin of the patient and surgeon. Surg
Infect (Larchmt). 2006;7(suppl 3):3–15.
16. Darouiche RO, Wall MJ, Jr, Itani KM, et al. Chlorhexidine-alcohol versus
povidone-iodine for surgical-site antisepsis. N Engl J Med. 2010;362:18–26.
17. Denton GW. Chlorhexidine. In: Block SS, ed. Disinfection, Sterilization,
and Preservation. 5th ed. Philadelphia, PA: Lippincott Williams & Wilkins;
2001:321–336.
18. Culligan PJ, Kubik K, Murphy M, et al. A randomized trial that compared
povidone iodine and chlorhexidine as antiseptics for vaginal hysterectomy. Am
J Obstet Gynecol. 2005;192:422–425.
19. Ostrander RV, Botte MJ, Brage ME. Efficacy of surgical preparation solutions
in foot and ankle surgery. J Bone Joint Surg Am. 2005;87:980–985.
20. Swenson BR, Hedrick TL, Metzger R, et al. Effects of preoperative skin preparation
on postoperative wound infection rates: a prospective study of 3 skin
preparation protocols. Infect Control Hosp Epidemiol. 2009;30:964–971. |
/1 
发表于 2012-3-15 08:56
发表于 2012-3-15 10:20
