- Email: [email protected]
The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery
IJG-08121; No of Pages 5 International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery Finn Egil Skjeldestad a,b,⁎, Jørgen V. Bjørnholt a, Jon M. Gran a,c, Hanne-Merete Erisken a a b c
Department of Infectious Disease Epidemiology, National Institute of Public Health, Oslo, Norway Department of Clinical Medicine, The University of Tromsø, The Arctic University of Norway, Tromsø, Norway Department of Biostatistics, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway
a r t i c l e
i n f o
Article history: Received 17 March 2014 Received in revised form 5 August 2014 Accepted 7 October 2014 Keywords: Antibiotic prophylaxis Cesarean delivery Guidelines Infection control Infection prevention Surgical-site infection
a b s t r a c t Objective: To evaluate the effect of Norwegian antibiotic prophylaxis guidelines on rates of superficial and deep surgical-site infections (SSIs) associated with cesarean delivery (CD). Methods: A cross-sectional study was conducted that analyzed the physician-diagnosed SSIs by regimen of antibiotic prophylaxis among women who underwent planned or emergency CD at one of 42 hospitals between January 1, 2008, and December 31, 2010. The antibiotic prophylaxis regimen was verified using a hospital survey, whereas guideline compliance was assessed as part of the mandatory Norwegian Surveillance System for Healthcare-Associated Infections. Results: Data for 4498 patients were used. Hospitals that practiced antibiotic prophylaxis for all CDs (n = 4) provided antibiotics more often in both emergency and planned CDs than did those that used this approach for emergency CDs only (n = 33) or had no written guidelines or used prophylaxis on indication only (n = 5) (P b 0.001). The provision of antibiotic prophylaxis for all cases of CD was associated with markedly lowered rates of superficial SSIs among planned CDs, whereas no differences in rates of deep SSIs were observed between the guidelines in either planned or emergency CDs. Conclusion: Hospitals that provided antibiotic prophylaxis to all women undergoing CD reported high compliance and had reduced rates of superficial SSIs among planned CDs. © 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
1. Introduction The provision of antibiotic prophylaxis to women undergoing cesarean delivery (CD) has substantially reduced the incidence of febrile morbidity, wound infection, postpartum endometritis, and complications owing to severe maternal infections following both planned and emergency surgery [1]. Since the first studies on this approach were reported in the 1970s, many hospitals have introduced antibiotic prophylaxis as standard care to prevent infectious morbidity after CD [2, 3]. National clinical guidelines [1,4,5] have provided recommendations about the correct use of antibiotic prophylaxis in terms of indications, regimen, and timing among women who require CD. In Norway, 17.0% of the 62 000–63 000 deliveries in 2008–2010 were by cesarean, among which 38.0% were planned procedures [6]. Clinical guidelines published in 1999 by the Norwegian Society of Obstetrics and Gynecology in collaboration with the Norwegian Medical Association covered antibiotic prophylaxis for emergency CD only [7]. However, these guidelines were revised in 2008 to include long duration of surgery and severe hemorrhage as indications for antibiotic ⁎ Corresponding author at: Department of Clinical Medicine, The University of Tromsø, The Arctic University of Norway, N-9037 Tromsø, Norway. Tel.: +47 95 20 71 96; fax: +47 77 62 64 21. E-mail address: [email protected] (F.E. Skjeldestad).
prophylaxis among women undergoing planned CD [8]. Ampicillin or a first-generation cephalosporin are usually the drugs of choice for antibiotic prophylaxis, although the dose and timing were not stated in the guidelines [8]. Since 2005, all Norwegian hospitals have been required to report annually to the Norwegian Surveillance System for Healthcare-Associated Infections (NOIS) [9] on the rates of surgical-site infections (SSIs) arising from CD and four other non-gynecologic surgical procedures between September 1 and November 30. NOIS is a mandatory audit based on a European adaptation of guidelines that were originally developed by the US Centers for Disease Control and Prevention (CDC) [10,11]. The aims of NOIS are to monitor, describe, and evaluate the incidence of SSIs; to study the effects of interventions that aim to control infections; and to identify outbreaks so that SSIs can be prevented [12]. This surveillance system is one of 14 national health registries regulated by specific laws in Norway. The aim of the present study was to evaluate the effect of using the Norwegian antibiotic prophylaxis guidelines on the rates of SSIs associated with CD. 2. Materials and methods A cross-sectional study was conducted at 42 Norwegian hospitals that perform more than 10 CDs in the 3-month annual reporting period
http://dx.doi.org/10.1016/j.ijgo.2014.08.018 0020-7292/© 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018
2
F.E. Skjeldestad et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
[13]. It was a quality assurance study done within the NOIS framework, and all patient information was de-identified [12]. According to Norwegian regulations, quality assurance studies are exempt from obtaining written informed consent from the participants [12]. As part of the present study, the 42 Norwegian hospitals were surveyed about their use of written guidelines on antibiotic prophylaxis during CD between November 30, 2009, and April 26, 2010. Nonresponders were sent e-mail reminders and telephoned to obtain complete information on guidelines for antibiotic prophylaxis from all maternity departments. Departments were divided into groups on the basis of guideline availability: those in group A had no written guidelines available and provided antibiotic prophylaxis only when indicated; in group B, the intervention was provided for women undergoing emergency CDs only; and in group C, the intervention was provided for all CDs. NOIS data were used for women who underwent planned or emergency CD at the 42 hospitals between September 1 and November 30 in 2008, 2009, and 2010. In NOIS, data are collected both electronically and manually from medical records by infection control personnel and transferred to a standardized case-report form. Information collected included the level of care (university hospital, regional hospital, or local hospitals), patient characteristics (age and sex), dates of hospital admission and discharge, type of surgery (planned or emergency), duration of surgery, regimen of antibiotic prophylaxis provided, the American Society of Anesthesiologists (ASA) score, contamination class, and type of infection. All infections are diagnosed by a physician and comprise superficial, deep incision, and organ and/or space infections [10]. SSIs were classified as overall, superficial, and deep infections (including organ and/or space infections). Patients had to have been followed up for 30 days after surgery. Hospitals that did not deliver data for antibiotic prophylaxis were excluded. Additionally, patients were excluded if any information about them was missing. The hospital-specific guideline information from the survey was merged with the NOIS data. This procedure created two levels of information within the dataset: (1) individual data on compliance with the administration of antibiotics to prevent SSIs after CD and (2) hospital data on guidelines for antibiotic prophylaxis. Statistical analysis of the data was conducted by applying logistic regression and multilevel logistic regression with hospital and level of care (university hospital, regional hospital, or local hospital) as multilevel variables. Univariate analyses with χ2 tests and logistic regression were performed using SPSS version 19 (IBM, Armonk, NY, USA). Multilevel analyses were performed using lme4 R version 2.15.0 (R Core Team/R Foundation for Statistical Computing, Vienna, Austria). P b 0.05 was considered statistically significant.
centers. The remaining two regional hospitals and two local hospitals practiced antibiotic prophylaxis in all CDs (group C). Patients in group A were younger than were those in group B, whereas patients in Group C were older (P b 0.01) (Table 1). More emergency surgical procedures were performed at group C hospitals than at group A and B hospitals (P b 0.01) (Table 1). Only nonsignificant between-group differences were detected for ASA score, contamination class and duration of surgery. Group C hospitals provided antibiotics more often in both emergency and planned CDs than did group A and B hospitals (P b 0.001) (Table 1). No difference was detected in the provision of antibiotics between groups A and B in emergency CDs (P = 0.13); however, the four group A hospitals with no written guidelines provided antibiotics more often to patients who had undergone planned CDs (131 [63.3%] of 207 patients) than did group B hospitals (331 [23.1%] of 1435; P b 0.001). A total of 264 (5.9%) women in the study population had SSIs (Table 2). The frequency of superficial infections was higher in group A than in the other two groups (P = 0.02). Only non-significant differences in frequencies of SSIs were observed across level of care, age, type of surgery, antibiotic prophylaxis provided, and ASA score (Table 2). Compared with contamination classes 1 and 2 combined, both superficial infections and deep infections were more frequent in contamination classes 3 and 4 (P b 0.05), and in surgery that lasted for more than 45 minutes compared to procedures of shorter duration (P b 0.05). Differences in maternal morbidity, clinical indications, and circumstances within the operating theater can be predisposing factors for infectious morbidity. Consequently, the present analysis was stratified by emergency CD to investigate the association between antibiotic prophylaxis guidelines and occurrence of SSIs. A logistic regression analysis of planned CD showed that overall and superficial SSIs were more likely in groups A and B than in group C, whereas no difference was seen in the rate of deep SSIs (Table 3). By contrast, no between-group differences were detected in the rates of overall, superficial, and deep SSIs arising from emergency CD (Table 3). Similar results were observed in multilevel analyses (Table 3). The explanatory variables (age, ASA score, contamination class, and duration of surgery) exerted minimal statistically insignificant confounding effects (data not shown). The effect of antibiotic prophylaxis guidelines on overall SSI was evenly distributed across categories of the other explanatory variables (no interaction; data not shown). An evaluation of patients who received antibiotic prophylaxis versus those who did not revealed no differences in the rates of overall, superficial, and deep SSIs in either the planned or emergency CD groups (data not shown). 4. Discussion
3. Results A total of 6602 CDs were reported to NOIS across the three periods. One university hospital did not deliver data on antibiotic prophylaxis for the 1027 patients who underwent CD, whereas another university hospital did not deliver data for 143 patients in 2008. Of the remaining 5432 patients, 585 (10.8%) had incomplete follow-up after surgery. Thus, 4847 patients were potentially eligible for inclusion in the analyses. After exclusion of patients with missing information on age (n = 12; 0.2%), ASA classification (n = 108; 2.2%), wound contamination (n = 29; 0.6%), antibiotic prophylaxis (n = 198; 4.1%), and duration of surgery (n = 2; b 0.1%), the final dataset comprised 4498 patients, equivalent to 92.8% of the 4847 patients originally deemed eligible for analysis. Four local hospitals had no written guidelines on antibiotic prophylaxis, whereas one regional hospital practiced antibiotic prophylaxis on the basis of indication only (group A). Antibiotic prophylaxis was practiced in emergency CD only in 32 hospitals (group B), including all six included university hospitals, six regional hospitals, and 20 local
The present study showed a statistically significant association between lowered risk for superficial SSIs, but not deep SSIs, after planned CD and with guidelines recommending the use of antibiotic prophylaxis for all CDs. However, no differences in likelihood of SSIs between groups were reported for emergency CDs, which could be attributable to the high use of antibiotic prophylaxis for these procedures in all three groups. Lack of antibiotic prophylaxis might partly explain the higher odds of infection in groups A and B. Nevertheless, there was no significant difference in likelihood of SSIs after a planned CD between groups A and B, despite the fact that use of antibiotic prophylaxis after planned CD in group A was more than twice that of group B. This conflicting finding suggests that there might be additional causes for superficial SSI other than lack of antibiotic prophylaxis in planned CDs. Compliance with guidelines for both planned and emergency CDs was higher among the group C hospitals than in those in group B. Therefore, compliance with guidelines seems to be high when they cover all patients. This finding suggests that it might be easier to convey a
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018
F.E. Skjeldestad et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
3
Table 1 Characteristics of the study population by use of guidelines on antibiotic prophylaxis at their treating hospital.a Variable Level of care received University Regional Local Age, y 13–29 30–34 35–46 CD Planned Emergency Antibiotic prophylaxis provided In planned CD In emergency CD ASA score 1 2 3 4 Contamination class 1 2 3 4 Duration of surgery, min 8–29 30–44 45–279
Patients from group A hospitals (n = 442)b
Patients from group B hospitals (n = 3429)c
Patients from group C hospitals (n = 627)d
Total (n = 4498)
0 217 (49.1) 225 (50.9)
1625 (47.4) 925 (27.0) 879 (25.6)
0 396 (63.2) 231 (36.8)
1625 (36.1) 1538 (34.2) 1335 (29.7)
213 (48.2) 104 (23.5) 125 (28.3)
1334 (38.9) 899 (26.2) 1196 (34.9)
222 (35.4) 190 (30.3) 215 (34.3)
1769 (39.3) 1193 (26.5) 1536 (34.1)
207 (46.8) 235 (53.2)
1435 (41.8) 1994 (58.2)
204 (32.5) 423 (67.5)
1846 (41.0) 2652 (59.0)
131 (63.3) 200 (85.1)
331 (23.1) 1767 (88.6)
202 (99.0) 415 (98.1)
664 (36.0) 2382 (89.8)
127 (28.7) 304 (68.8) 11 (2.5) 0
1488 (43.4) 1836 (53.5) 100 (2.9) 5 (0.1)
334 (53.3) 280 (44.7) 12 (1.9) 1 (0.2)
1949 (43.3) 2420 (53.8) 123 (2.7) 6 (0.1)
253 (57.2) 184 (41.6) 4 (0.9) 1 (0.2)
2277 (66.4) 1083 (31.6) 66 (1.9) 3 (0.1)
415 (66.2) 192 (30.6) 20 (3.2) 0
2945 (65.5) 1459 (32.4) 90 (2.0) 4 (0.1)
149 (33.7) 198 (44.8) 95 (21.5)
1062 (31.0) 1506 (43.9) 861 (25.1)
209 (33.3) 277 (44.2) 141 (22.5)
1420 (31.6) 1981 (44.0) 1097 (24.4)
Abbreviations: CD, cesarean delivery; ASA, American Society of Anesthesiologists. a Values given as number (percentage). b No written guidelines on antibiotic prophylaxis/use on indication only. c Use of antibiotic prophylaxis in emergency CD only. d Use of antibiotic prophylaxis in all CDs.
guideline covering all patients to staff members involved and to organize a quality assurance system within the operating theater to fulfill the preventive intervention than it is to consider cases individually. Among hospital departments where no written guidelines are available on the use of antibiotic prophylaxis, the practice will be driven by the opinion of the individual obstetrician. A survey of maternal–fetal medicine physicians in the USA [14] found that only 2 (0.8%) of 245 respondents did not use routine antibiotics during CD. Approximately 95% of the physicians who used antibiotic prophylaxis stated that their drug of choice was either a first- or second-generation cephalosporin; 85% administered antibiotics before incision, in line with the recommendations of the American College of Obstetricians and Gynecologists [15]. In the present study, 32 of the 42 Norwegian hospital departments practiced the guideline on antibiotic prophylaxis in emergency CDs only, without preferences for timing, as proposed by the Norwegian Society of Obstetrics and Gynecology [8]. Over the present study period, more than 95% of all Norwegian hospitals performing CD reported to NOIS [9]. All SSIs are diagnosed by physicians and most cases are validated against source data according to CDC criteria by hospital infection-control personnel before the data are transferred to NOIS. Because most SSIs are diagnosed after discharge [9], a complete active follow-up (as documented in the present study) is important for reliable estimates for SSIs within 30 days of surgery. The overall SSI rate reported in the present study was similar to that reported for 1999–2001 by a US tertiary hospital, which used CDC diagnostic criteria [16]. In addition, the present findings were similar to those reported by the US National Nosocomial Infections Surveillance System to the CDC from January 1992 through June 2004 [17]. However, comparisons of the rates of SSIs across studies are complicated by differences in study populations, definitions of infection, lack of
(or differences in) duration and completeness of postdischarge followup data, and use of antibiotic prophylaxis. Consequently, rates of SSIs associated with CD vary considerably [1] and must be interpreted with caution. The purpose of antibiotic prophylaxis is not to sterilize tissue, but to reduce the intraoperative load of microorganisms to a level that the patient’s defense systems have the capacity to overcome. The average duration of CD is 30–40 minutes [18]. Therefore, it is important that the serum concentration of antibiotic is established before dermal incision to ensure sufficient tissue concentrations are reached by the time of maximum trauma when delivery occurs through the uterotomy and abdominal wall shortly after surgery starts. Except for cases in which rupture of the membranes persists for hours, CD can be considered clean contaminated surgery (class 2) if the lower uterine segment is not reached during surgery. If the obstetrician reaches the lower uterine segment, the surgery is contaminated (class 3). As expected, the present study found an increased incidence of deep infections in emergency CDs versus planned CDs but no difference in superficial infections. In some emergency CDs, there might be limited time available for personal hygiene (as the woman has often been in labor for hours), for surgical scrubbing among team members, and for adequate lining of the operation table [11]. Furthermore, rapid entry through the abdominal wall and uterus can create bleeding problems that are not easily resolved. Large tears in the corners of the uterotomy might challenge the reconstructive skills of the obstetrician. All these issues could lead to contamination and tissue trauma in complicated surgery that predisposes an individual to infection after the effects of a single dose of antibiotic prophylaxis have worn off (the half-life of a cephalosporin is 30–60 minutes). A limitation of NOIS is the fact that this audit does not capture any variables for complicated surgery other than duration of the procedure. In addition, the low incidence of deep infections might explain why no
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018
4
F.E. Skjeldestad et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
Table 2 Characteristics of the study population by type of surgical-site infection.a Variable
Total number of patients
Total for study population Hospital Group Ab Group Bc Group Cd Level of care University Regional Local Age, y 13–29 30–34 35–46 CD Planned Emergency Antibiotic prophylaxis provided in planned CD Yes No Antibiotic prophylaxis provided in emergency CD Yes No ASA score 1 2 3 4 Contamination class 1 2 3 4 Duration of surgery, min 8–29 30–44 45–279
Type of surgical-site infection None
Superficial
Deep
4498
4234 (94.1)
184 (4.1)
80 (1.8)
442 3429 627
406 (91.9) 3235 (94.3) 593 (94.6)
31 (7.0) 130 (3.8) 23 (3.7)
5 (1.1) 64 (1.9) 11 (1.8)
1625 1538 1335
1521 (93.6) 1452 (94.4) 1261 (94.5)
79 (4.9) 54 (3.5) 51 (3.8)
25 (1.5) 32 (2.1) 23 (1.7)
1769 1536 1193
1664 (94.1) 1454 (94.7) 1116 (93.5)
63 (3.6) 61 (4.0) 60 (5.0)
42 (2.4) 21 (1.4) 17 (1.4)
1846 2652
1748 (94.7) 2486 (93.7)
75 (4.1) 109 (4.1)
23 (1.2) 57 (2.1)
664 1182
628 (94.6) 1120 (94.8)
27 (4.1) 48 (4.1)
9 (1.4) 14 (1.2)
2382 270
2231 (93.7) 255 (94.4)
97 (4.1) 12 (4.4)
54 (2.3) 3 (1.1)
1949 2420 123 6
1835 (94.2) 2280 (94.2) 113 (91.9) 6 (100.0)
79 (4.1) 96 (4.0) 9 (7.3) 0
35 (1.8) 44 (1.8) 1 (0.8) 0
2945 1459 90 4
2779 (94.4) 1372 (94.0) 80 (88.9) 3 (75.0)
118 (4.0) 59 (4.0) 6 (6.7) 1 (25.0)
48 (1.6) 28 (1.9) 4 (4.4) 0
1420 1981 1097
1347 (94.9) 1868 (94.3) 1019 (92.9)
49 (3.5) 84 (4.2) 51 (4.6)
24 (1.7) 29 (1.5) 27 (2.5)
Abbreviations: CD, cesarean delivery; ASA, American Society of Anesthesiologists. a Values given as number or number (percentage). b No written guidelines on antibiotic prophylaxis/use on indication only. c Use of antibiotic prophylaxis in emergency CD only. d Use of antibiotic prophylaxis in all CDs.
Table 3 The association between guidelines on antibiotic prophylaxis and type of SSI.a Variable
Overall SSI Eligible for analysis Prevalence Group Ab Group Bc Group Cd Superficial SSI Eligible for analysis Prevalence Group Ab Group Bc Group Cd Deep SSI Eligible for analysis Prevalence Group Ab Group Bc Group Cd
Logistic regression
Multilevel analysis
Planned CD
Emergency CD
Planned CD
Emergency CD
1846 5.3 4.8 (1.6–14.3) 2.8 (1.0–7.7) 1.0
2652 6.3 1.1 (0.6–2.0) 0.8 (0.5–1.3) 1.0
1846 5.3 4.4 (1.3–15.2) 2.9 (0.9–8.6) 1.0
2652 6.3 1.3 (0.4–4.1) 1.1 (0.4–2.6) 1.0
1823 4.1 5.3 (1.5–18.6) 2.8 (0.9–9.0) 1.0
2595 4.2 1.5 (0.7–2.9) 0.8 (0.5–1.3) 1.0
1823 4.1 4.1 (0.8–21.9) 2.8 (0.6–12.1) 1.0
2595 4.2 1.6 (0.5–4.5) 0.8 (0.4–1.8) 1.0
1771 1.3 3.2 (0.3–30.8) 2.8 (0.3–21.0) 1.0
2543 2.2 0.4 (0.1–1.7) 0.9 (0.5–1.9) 1.0
1771 1.3 3.2 (0.3–30.8) 2.8 (0.4–21.0) 1.0
2543 2.2 0.5 (0.0–9.1) 1.8 (0.3–12.5) 1.0
Abbreviations: SSI, surgical-site infection; CD, cesarean delivery. a Values given as number, percentage, or odds ratio (95% confidence interval). b No written guidelines on antibiotic prophylaxis/use on indication only. c Use of antibiotic prophylaxis in emergency CD only. d Use of antibiotic prophylaxis in all CDs.
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018
F.E. Skjeldestad et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
difference was found between the three groups with regard to the effect of antibiotic prophylaxis in this type of SSI. The present study has a few strengths: prospective data collection was done using a national surveillance system with a very high hospital participation rate; complete data are available for 30 days of follow-up; and most diagnoses of SSIs were validated by trained infection control personnel. Furthermore, the availability of individual data on the provision on antibiotic prophylaxis could be considered a strength of the current analysis. Limitations of the present study included lack of validation of the diagnosis of SSIs or whether there is a differential under-reporting or misclassification between superficial and deep infections. For example, in 2010, one hospital in group A changed its protocol from giving antibiotic prophylaxis on indication only to provision of antibiotic prophylaxis for all CDs. However, this case seems to be isolated; none of the other hospitals involved in the present investigation reported a change in guideline during the study period. In conclusion, the findings of the present study indicate that hospital departments that provided antibiotic prophylaxis for all CDs had higher guideline compliance in both planned and emergency procedures than did those that either practiced antibiotic prophylaxis in emergency CDs only or had no written guidelines. High levels of compliance with guidelines might partly explain the reduced incidence of superficial SSIs among planned CDs; however, compliance had no effect on superficial SSI among emergency CDs or the incidence of deep SSIs. Conflict of interest The authors have no conflicts of interest. References [1] Smaill FM, Gyte GM. Antibiotic prophylaxis versus no prophylaxis for preventing infection after cesarean section. Cochrane Database Syst Rev 2010;1:CD007482. [2] Gall SA. The efficacy of prophylactic antibiotics in cesarean section. Am J Obstet Gynecol 1979;134:506–11. [3] Dillon WP, Seigel MS, Lele AS, O’Leary JA. Evaluation of cefoxitin prophylaxis for cesarean section. Int J Gynecol Obstet 1981;19(2):133–9.
5
[4] American College of Obstetricians and Gynecologists. ACOG practice bulleting number 47, October 2003: Prophylactic antibiotics in labor and delivery. Obstet Gynecol 2003;102(4):875–82. [5] van Schalkwyk J, Van Eyk N, Society of Obstetricians and Gynaecologists of Canada Infectious Diseases Committee. Antibiotic prophylaxis in obstetric procedures. J Obstet Gynaecol Can 2010;32(9):878–92. [6] National Institute of Public Health. Medical Birth Register. http://www.fhi.no/ helseregistre/medisinsk-fodselsregister. Published November 21, 2012. Accessed September 26, 2014. [7] Lichtenberg SM, Bungum L, Lande E. Cesarean section. In: Dalaker K, Berle EJ, editors. Clinical Guidelines in Obstetrics 1998 (in Norwegian). Oslo: The Norwegian Society of Obstetrics and Gynecology, The Norwegian Medical Association; 1998. [8] Häger R, Kolås T, Sviggum O, Novakovic Z. Cesarean Section. (in Norwegian) http:// legeforeningen.no/Fagmed/Norsk-gynekologisk-forening/Veiledere/veileder-ifodselshjelp-2008/. Published September 27, 2008. Accessed September 26, 2014. [9] Løwer HL, Eriksen HM, Aavitsland P, Skjeldestad FE. Methodology of the Norwegian Surveillance System for Healthcare-Associated Infections: the value of a mandatory system, automated data collection, and active postdischarge surveillance. Am J Infect Control 2013;41(7):591–6. [10] Hospital in Europe Link for Infection Control through Surveillance (HELICS), European Centre for Disease Prevention and Control. Surveillance of Surgical Site Infections: Master Protocol. http://www.ecdc.europa.eu/en/activities/surveillance/ HAI/Documents/0409_IPSE_SSI_protocol.pdf. Published September 2004. Accessed July 16, 2014. [11] Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1999;20(4):250–78. [12] Lovdata. Regulations on the Norwegian Surveillance System for antibiotic use and healthcare-associated infections (NOIS registry regulations). http://lovdata.no/ dokument/SF/forskrift/2005-06-17-611?q=Helseregister+forskriften. Published June 21, 2005. Accessed July 16, 2014. [13] Eriksen HM, Sæther AR, Økland I, Langen E, Sandness Y, Bødtker A, et al. Antibiotics prophylaxis in connection with caesarean section–guidelines at Norwegian maternity departments. Tidsskr Nor Laegeforen 2011;131(23):2355–8. [14] Doss AE, Davidson JD, Cliver SP, Wetta LA, Andrews WW, Tita AT. Antibiotic prophylaxis for cesarean delivery: survey of maternal-fetal medicine physicians in the U.S. J Matern Fetal Neonatal Med 2012;25(8):1264–6. [15] Committee opinion no. 465: antimicrobial prophylaxis for cesarean delivery: timing of administration. Obstet Gynecol 2010;116(3):791–2. [16] Olsen MA, Butler AM, Willers DM, Devkota P, Gross GA, Fraser VJ. Risk factors for surgical site infection after low transverse cesarean section. Infect Control Hosp Epidemiol 2008;29(6):477–86. [17] National Nosocomial Infections Surveillance System. National Nosocomial Infections Surveillance (NNIS) System Report, data summary from January 1992 through June 2004, issued October 2004. Am J Infect Control 2004;32(8):470–85. [18] Hofmeyr JG, Novikova N, Mathai M, Shah A. Techniques for cesarean section. Am J Obstet Gynecol 2009;201(5):431–44.
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery Finn Egil Skjeldestad a,b,⁎, Jørgen V. Bjørnholt a, Jon M. Gran a,c, Hanne-Merete Erisken a a b c
Department of Infectious Disease Epidemiology, National Institute of Public Health, Oslo, Norway Department of Clinical Medicine, The University of Tromsø, The Arctic University of Norway, Tromsø, Norway Department of Biostatistics, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway
a r t i c l e
i n f o
Article history: Received 17 March 2014 Received in revised form 5 August 2014 Accepted 7 October 2014 Keywords: Antibiotic prophylaxis Cesarean delivery Guidelines Infection control Infection prevention Surgical-site infection
a b s t r a c t Objective: To evaluate the effect of Norwegian antibiotic prophylaxis guidelines on rates of superficial and deep surgical-site infections (SSIs) associated with cesarean delivery (CD). Methods: A cross-sectional study was conducted that analyzed the physician-diagnosed SSIs by regimen of antibiotic prophylaxis among women who underwent planned or emergency CD at one of 42 hospitals between January 1, 2008, and December 31, 2010. The antibiotic prophylaxis regimen was verified using a hospital survey, whereas guideline compliance was assessed as part of the mandatory Norwegian Surveillance System for Healthcare-Associated Infections. Results: Data for 4498 patients were used. Hospitals that practiced antibiotic prophylaxis for all CDs (n = 4) provided antibiotics more often in both emergency and planned CDs than did those that used this approach for emergency CDs only (n = 33) or had no written guidelines or used prophylaxis on indication only (n = 5) (P b 0.001). The provision of antibiotic prophylaxis for all cases of CD was associated with markedly lowered rates of superficial SSIs among planned CDs, whereas no differences in rates of deep SSIs were observed between the guidelines in either planned or emergency CDs. Conclusion: Hospitals that provided antibiotic prophylaxis to all women undergoing CD reported high compliance and had reduced rates of superficial SSIs among planned CDs. © 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
1. Introduction The provision of antibiotic prophylaxis to women undergoing cesarean delivery (CD) has substantially reduced the incidence of febrile morbidity, wound infection, postpartum endometritis, and complications owing to severe maternal infections following both planned and emergency surgery [1]. Since the first studies on this approach were reported in the 1970s, many hospitals have introduced antibiotic prophylaxis as standard care to prevent infectious morbidity after CD [2, 3]. National clinical guidelines [1,4,5] have provided recommendations about the correct use of antibiotic prophylaxis in terms of indications, regimen, and timing among women who require CD. In Norway, 17.0% of the 62 000–63 000 deliveries in 2008–2010 were by cesarean, among which 38.0% were planned procedures [6]. Clinical guidelines published in 1999 by the Norwegian Society of Obstetrics and Gynecology in collaboration with the Norwegian Medical Association covered antibiotic prophylaxis for emergency CD only [7]. However, these guidelines were revised in 2008 to include long duration of surgery and severe hemorrhage as indications for antibiotic ⁎ Corresponding author at: Department of Clinical Medicine, The University of Tromsø, The Arctic University of Norway, N-9037 Tromsø, Norway. Tel.: +47 95 20 71 96; fax: +47 77 62 64 21. E-mail address: [email protected] (F.E. Skjeldestad).
prophylaxis among women undergoing planned CD [8]. Ampicillin or a first-generation cephalosporin are usually the drugs of choice for antibiotic prophylaxis, although the dose and timing were not stated in the guidelines [8]. Since 2005, all Norwegian hospitals have been required to report annually to the Norwegian Surveillance System for Healthcare-Associated Infections (NOIS) [9] on the rates of surgical-site infections (SSIs) arising from CD and four other non-gynecologic surgical procedures between September 1 and November 30. NOIS is a mandatory audit based on a European adaptation of guidelines that were originally developed by the US Centers for Disease Control and Prevention (CDC) [10,11]. The aims of NOIS are to monitor, describe, and evaluate the incidence of SSIs; to study the effects of interventions that aim to control infections; and to identify outbreaks so that SSIs can be prevented [12]. This surveillance system is one of 14 national health registries regulated by specific laws in Norway. The aim of the present study was to evaluate the effect of using the Norwegian antibiotic prophylaxis guidelines on the rates of SSIs associated with CD. 2. Materials and methods A cross-sectional study was conducted at 42 Norwegian hospitals that perform more than 10 CDs in the 3-month annual reporting period
http://dx.doi.org/10.1016/j.ijgo.2014.08.018 0020-7292/© 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018
2
F.E. Skjeldestad et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
[13]. It was a quality assurance study done within the NOIS framework, and all patient information was de-identified [12]. According to Norwegian regulations, quality assurance studies are exempt from obtaining written informed consent from the participants [12]. As part of the present study, the 42 Norwegian hospitals were surveyed about their use of written guidelines on antibiotic prophylaxis during CD between November 30, 2009, and April 26, 2010. Nonresponders were sent e-mail reminders and telephoned to obtain complete information on guidelines for antibiotic prophylaxis from all maternity departments. Departments were divided into groups on the basis of guideline availability: those in group A had no written guidelines available and provided antibiotic prophylaxis only when indicated; in group B, the intervention was provided for women undergoing emergency CDs only; and in group C, the intervention was provided for all CDs. NOIS data were used for women who underwent planned or emergency CD at the 42 hospitals between September 1 and November 30 in 2008, 2009, and 2010. In NOIS, data are collected both electronically and manually from medical records by infection control personnel and transferred to a standardized case-report form. Information collected included the level of care (university hospital, regional hospital, or local hospitals), patient characteristics (age and sex), dates of hospital admission and discharge, type of surgery (planned or emergency), duration of surgery, regimen of antibiotic prophylaxis provided, the American Society of Anesthesiologists (ASA) score, contamination class, and type of infection. All infections are diagnosed by a physician and comprise superficial, deep incision, and organ and/or space infections [10]. SSIs were classified as overall, superficial, and deep infections (including organ and/or space infections). Patients had to have been followed up for 30 days after surgery. Hospitals that did not deliver data for antibiotic prophylaxis were excluded. Additionally, patients were excluded if any information about them was missing. The hospital-specific guideline information from the survey was merged with the NOIS data. This procedure created two levels of information within the dataset: (1) individual data on compliance with the administration of antibiotics to prevent SSIs after CD and (2) hospital data on guidelines for antibiotic prophylaxis. Statistical analysis of the data was conducted by applying logistic regression and multilevel logistic regression with hospital and level of care (university hospital, regional hospital, or local hospital) as multilevel variables. Univariate analyses with χ2 tests and logistic regression were performed using SPSS version 19 (IBM, Armonk, NY, USA). Multilevel analyses were performed using lme4 R version 2.15.0 (R Core Team/R Foundation for Statistical Computing, Vienna, Austria). P b 0.05 was considered statistically significant.
centers. The remaining two regional hospitals and two local hospitals practiced antibiotic prophylaxis in all CDs (group C). Patients in group A were younger than were those in group B, whereas patients in Group C were older (P b 0.01) (Table 1). More emergency surgical procedures were performed at group C hospitals than at group A and B hospitals (P b 0.01) (Table 1). Only nonsignificant between-group differences were detected for ASA score, contamination class and duration of surgery. Group C hospitals provided antibiotics more often in both emergency and planned CDs than did group A and B hospitals (P b 0.001) (Table 1). No difference was detected in the provision of antibiotics between groups A and B in emergency CDs (P = 0.13); however, the four group A hospitals with no written guidelines provided antibiotics more often to patients who had undergone planned CDs (131 [63.3%] of 207 patients) than did group B hospitals (331 [23.1%] of 1435; P b 0.001). A total of 264 (5.9%) women in the study population had SSIs (Table 2). The frequency of superficial infections was higher in group A than in the other two groups (P = 0.02). Only non-significant differences in frequencies of SSIs were observed across level of care, age, type of surgery, antibiotic prophylaxis provided, and ASA score (Table 2). Compared with contamination classes 1 and 2 combined, both superficial infections and deep infections were more frequent in contamination classes 3 and 4 (P b 0.05), and in surgery that lasted for more than 45 minutes compared to procedures of shorter duration (P b 0.05). Differences in maternal morbidity, clinical indications, and circumstances within the operating theater can be predisposing factors for infectious morbidity. Consequently, the present analysis was stratified by emergency CD to investigate the association between antibiotic prophylaxis guidelines and occurrence of SSIs. A logistic regression analysis of planned CD showed that overall and superficial SSIs were more likely in groups A and B than in group C, whereas no difference was seen in the rate of deep SSIs (Table 3). By contrast, no between-group differences were detected in the rates of overall, superficial, and deep SSIs arising from emergency CD (Table 3). Similar results were observed in multilevel analyses (Table 3). The explanatory variables (age, ASA score, contamination class, and duration of surgery) exerted minimal statistically insignificant confounding effects (data not shown). The effect of antibiotic prophylaxis guidelines on overall SSI was evenly distributed across categories of the other explanatory variables (no interaction; data not shown). An evaluation of patients who received antibiotic prophylaxis versus those who did not revealed no differences in the rates of overall, superficial, and deep SSIs in either the planned or emergency CD groups (data not shown). 4. Discussion
3. Results A total of 6602 CDs were reported to NOIS across the three periods. One university hospital did not deliver data on antibiotic prophylaxis for the 1027 patients who underwent CD, whereas another university hospital did not deliver data for 143 patients in 2008. Of the remaining 5432 patients, 585 (10.8%) had incomplete follow-up after surgery. Thus, 4847 patients were potentially eligible for inclusion in the analyses. After exclusion of patients with missing information on age (n = 12; 0.2%), ASA classification (n = 108; 2.2%), wound contamination (n = 29; 0.6%), antibiotic prophylaxis (n = 198; 4.1%), and duration of surgery (n = 2; b 0.1%), the final dataset comprised 4498 patients, equivalent to 92.8% of the 4847 patients originally deemed eligible for analysis. Four local hospitals had no written guidelines on antibiotic prophylaxis, whereas one regional hospital practiced antibiotic prophylaxis on the basis of indication only (group A). Antibiotic prophylaxis was practiced in emergency CD only in 32 hospitals (group B), including all six included university hospitals, six regional hospitals, and 20 local
The present study showed a statistically significant association between lowered risk for superficial SSIs, but not deep SSIs, after planned CD and with guidelines recommending the use of antibiotic prophylaxis for all CDs. However, no differences in likelihood of SSIs between groups were reported for emergency CDs, which could be attributable to the high use of antibiotic prophylaxis for these procedures in all three groups. Lack of antibiotic prophylaxis might partly explain the higher odds of infection in groups A and B. Nevertheless, there was no significant difference in likelihood of SSIs after a planned CD between groups A and B, despite the fact that use of antibiotic prophylaxis after planned CD in group A was more than twice that of group B. This conflicting finding suggests that there might be additional causes for superficial SSI other than lack of antibiotic prophylaxis in planned CDs. Compliance with guidelines for both planned and emergency CDs was higher among the group C hospitals than in those in group B. Therefore, compliance with guidelines seems to be high when they cover all patients. This finding suggests that it might be easier to convey a
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018
F.E. Skjeldestad et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
3
Table 1 Characteristics of the study population by use of guidelines on antibiotic prophylaxis at their treating hospital.a Variable Level of care received University Regional Local Age, y 13–29 30–34 35–46 CD Planned Emergency Antibiotic prophylaxis provided In planned CD In emergency CD ASA score 1 2 3 4 Contamination class 1 2 3 4 Duration of surgery, min 8–29 30–44 45–279
Patients from group A hospitals (n = 442)b
Patients from group B hospitals (n = 3429)c
Patients from group C hospitals (n = 627)d
Total (n = 4498)
0 217 (49.1) 225 (50.9)
1625 (47.4) 925 (27.0) 879 (25.6)
0 396 (63.2) 231 (36.8)
1625 (36.1) 1538 (34.2) 1335 (29.7)
213 (48.2) 104 (23.5) 125 (28.3)
1334 (38.9) 899 (26.2) 1196 (34.9)
222 (35.4) 190 (30.3) 215 (34.3)
1769 (39.3) 1193 (26.5) 1536 (34.1)
207 (46.8) 235 (53.2)
1435 (41.8) 1994 (58.2)
204 (32.5) 423 (67.5)
1846 (41.0) 2652 (59.0)
131 (63.3) 200 (85.1)
331 (23.1) 1767 (88.6)
202 (99.0) 415 (98.1)
664 (36.0) 2382 (89.8)
127 (28.7) 304 (68.8) 11 (2.5) 0
1488 (43.4) 1836 (53.5) 100 (2.9) 5 (0.1)
334 (53.3) 280 (44.7) 12 (1.9) 1 (0.2)
1949 (43.3) 2420 (53.8) 123 (2.7) 6 (0.1)
253 (57.2) 184 (41.6) 4 (0.9) 1 (0.2)
2277 (66.4) 1083 (31.6) 66 (1.9) 3 (0.1)
415 (66.2) 192 (30.6) 20 (3.2) 0
2945 (65.5) 1459 (32.4) 90 (2.0) 4 (0.1)
149 (33.7) 198 (44.8) 95 (21.5)
1062 (31.0) 1506 (43.9) 861 (25.1)
209 (33.3) 277 (44.2) 141 (22.5)
1420 (31.6) 1981 (44.0) 1097 (24.4)
Abbreviations: CD, cesarean delivery; ASA, American Society of Anesthesiologists. a Values given as number (percentage). b No written guidelines on antibiotic prophylaxis/use on indication only. c Use of antibiotic prophylaxis in emergency CD only. d Use of antibiotic prophylaxis in all CDs.
guideline covering all patients to staff members involved and to organize a quality assurance system within the operating theater to fulfill the preventive intervention than it is to consider cases individually. Among hospital departments where no written guidelines are available on the use of antibiotic prophylaxis, the practice will be driven by the opinion of the individual obstetrician. A survey of maternal–fetal medicine physicians in the USA [14] found that only 2 (0.8%) of 245 respondents did not use routine antibiotics during CD. Approximately 95% of the physicians who used antibiotic prophylaxis stated that their drug of choice was either a first- or second-generation cephalosporin; 85% administered antibiotics before incision, in line with the recommendations of the American College of Obstetricians and Gynecologists [15]. In the present study, 32 of the 42 Norwegian hospital departments practiced the guideline on antibiotic prophylaxis in emergency CDs only, without preferences for timing, as proposed by the Norwegian Society of Obstetrics and Gynecology [8]. Over the present study period, more than 95% of all Norwegian hospitals performing CD reported to NOIS [9]. All SSIs are diagnosed by physicians and most cases are validated against source data according to CDC criteria by hospital infection-control personnel before the data are transferred to NOIS. Because most SSIs are diagnosed after discharge [9], a complete active follow-up (as documented in the present study) is important for reliable estimates for SSIs within 30 days of surgery. The overall SSI rate reported in the present study was similar to that reported for 1999–2001 by a US tertiary hospital, which used CDC diagnostic criteria [16]. In addition, the present findings were similar to those reported by the US National Nosocomial Infections Surveillance System to the CDC from January 1992 through June 2004 [17]. However, comparisons of the rates of SSIs across studies are complicated by differences in study populations, definitions of infection, lack of
(or differences in) duration and completeness of postdischarge followup data, and use of antibiotic prophylaxis. Consequently, rates of SSIs associated with CD vary considerably [1] and must be interpreted with caution. The purpose of antibiotic prophylaxis is not to sterilize tissue, but to reduce the intraoperative load of microorganisms to a level that the patient’s defense systems have the capacity to overcome. The average duration of CD is 30–40 minutes [18]. Therefore, it is important that the serum concentration of antibiotic is established before dermal incision to ensure sufficient tissue concentrations are reached by the time of maximum trauma when delivery occurs through the uterotomy and abdominal wall shortly after surgery starts. Except for cases in which rupture of the membranes persists for hours, CD can be considered clean contaminated surgery (class 2) if the lower uterine segment is not reached during surgery. If the obstetrician reaches the lower uterine segment, the surgery is contaminated (class 3). As expected, the present study found an increased incidence of deep infections in emergency CDs versus planned CDs but no difference in superficial infections. In some emergency CDs, there might be limited time available for personal hygiene (as the woman has often been in labor for hours), for surgical scrubbing among team members, and for adequate lining of the operation table [11]. Furthermore, rapid entry through the abdominal wall and uterus can create bleeding problems that are not easily resolved. Large tears in the corners of the uterotomy might challenge the reconstructive skills of the obstetrician. All these issues could lead to contamination and tissue trauma in complicated surgery that predisposes an individual to infection after the effects of a single dose of antibiotic prophylaxis have worn off (the half-life of a cephalosporin is 30–60 minutes). A limitation of NOIS is the fact that this audit does not capture any variables for complicated surgery other than duration of the procedure. In addition, the low incidence of deep infections might explain why no
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018
4
F.E. Skjeldestad et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
Table 2 Characteristics of the study population by type of surgical-site infection.a Variable
Total number of patients
Total for study population Hospital Group Ab Group Bc Group Cd Level of care University Regional Local Age, y 13–29 30–34 35–46 CD Planned Emergency Antibiotic prophylaxis provided in planned CD Yes No Antibiotic prophylaxis provided in emergency CD Yes No ASA score 1 2 3 4 Contamination class 1 2 3 4 Duration of surgery, min 8–29 30–44 45–279
Type of surgical-site infection None
Superficial
Deep
4498
4234 (94.1)
184 (4.1)
80 (1.8)
442 3429 627
406 (91.9) 3235 (94.3) 593 (94.6)
31 (7.0) 130 (3.8) 23 (3.7)
5 (1.1) 64 (1.9) 11 (1.8)
1625 1538 1335
1521 (93.6) 1452 (94.4) 1261 (94.5)
79 (4.9) 54 (3.5) 51 (3.8)
25 (1.5) 32 (2.1) 23 (1.7)
1769 1536 1193
1664 (94.1) 1454 (94.7) 1116 (93.5)
63 (3.6) 61 (4.0) 60 (5.0)
42 (2.4) 21 (1.4) 17 (1.4)
1846 2652
1748 (94.7) 2486 (93.7)
75 (4.1) 109 (4.1)
23 (1.2) 57 (2.1)
664 1182
628 (94.6) 1120 (94.8)
27 (4.1) 48 (4.1)
9 (1.4) 14 (1.2)
2382 270
2231 (93.7) 255 (94.4)
97 (4.1) 12 (4.4)
54 (2.3) 3 (1.1)
1949 2420 123 6
1835 (94.2) 2280 (94.2) 113 (91.9) 6 (100.0)
79 (4.1) 96 (4.0) 9 (7.3) 0
35 (1.8) 44 (1.8) 1 (0.8) 0
2945 1459 90 4
2779 (94.4) 1372 (94.0) 80 (88.9) 3 (75.0)
118 (4.0) 59 (4.0) 6 (6.7) 1 (25.0)
48 (1.6) 28 (1.9) 4 (4.4) 0
1420 1981 1097
1347 (94.9) 1868 (94.3) 1019 (92.9)
49 (3.5) 84 (4.2) 51 (4.6)
24 (1.7) 29 (1.5) 27 (2.5)
Abbreviations: CD, cesarean delivery; ASA, American Society of Anesthesiologists. a Values given as number or number (percentage). b No written guidelines on antibiotic prophylaxis/use on indication only. c Use of antibiotic prophylaxis in emergency CD only. d Use of antibiotic prophylaxis in all CDs.
Table 3 The association between guidelines on antibiotic prophylaxis and type of SSI.a Variable
Overall SSI Eligible for analysis Prevalence Group Ab Group Bc Group Cd Superficial SSI Eligible for analysis Prevalence Group Ab Group Bc Group Cd Deep SSI Eligible for analysis Prevalence Group Ab Group Bc Group Cd
Logistic regression
Multilevel analysis
Planned CD
Emergency CD
Planned CD
Emergency CD
1846 5.3 4.8 (1.6–14.3) 2.8 (1.0–7.7) 1.0
2652 6.3 1.1 (0.6–2.0) 0.8 (0.5–1.3) 1.0
1846 5.3 4.4 (1.3–15.2) 2.9 (0.9–8.6) 1.0
2652 6.3 1.3 (0.4–4.1) 1.1 (0.4–2.6) 1.0
1823 4.1 5.3 (1.5–18.6) 2.8 (0.9–9.0) 1.0
2595 4.2 1.5 (0.7–2.9) 0.8 (0.5–1.3) 1.0
1823 4.1 4.1 (0.8–21.9) 2.8 (0.6–12.1) 1.0
2595 4.2 1.6 (0.5–4.5) 0.8 (0.4–1.8) 1.0
1771 1.3 3.2 (0.3–30.8) 2.8 (0.3–21.0) 1.0
2543 2.2 0.4 (0.1–1.7) 0.9 (0.5–1.9) 1.0
1771 1.3 3.2 (0.3–30.8) 2.8 (0.4–21.0) 1.0
2543 2.2 0.5 (0.0–9.1) 1.8 (0.3–12.5) 1.0
Abbreviations: SSI, surgical-site infection; CD, cesarean delivery. a Values given as number, percentage, or odds ratio (95% confidence interval). b No written guidelines on antibiotic prophylaxis/use on indication only. c Use of antibiotic prophylaxis in emergency CD only. d Use of antibiotic prophylaxis in all CDs.
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018
F.E. Skjeldestad et al. / International Journal of Gynecology and Obstetrics xxx (2014) xxx–xxx
difference was found between the three groups with regard to the effect of antibiotic prophylaxis in this type of SSI. The present study has a few strengths: prospective data collection was done using a national surveillance system with a very high hospital participation rate; complete data are available for 30 days of follow-up; and most diagnoses of SSIs were validated by trained infection control personnel. Furthermore, the availability of individual data on the provision on antibiotic prophylaxis could be considered a strength of the current analysis. Limitations of the present study included lack of validation of the diagnosis of SSIs or whether there is a differential under-reporting or misclassification between superficial and deep infections. For example, in 2010, one hospital in group A changed its protocol from giving antibiotic prophylaxis on indication only to provision of antibiotic prophylaxis for all CDs. However, this case seems to be isolated; none of the other hospitals involved in the present investigation reported a change in guideline during the study period. In conclusion, the findings of the present study indicate that hospital departments that provided antibiotic prophylaxis for all CDs had higher guideline compliance in both planned and emergency procedures than did those that either practiced antibiotic prophylaxis in emergency CDs only or had no written guidelines. High levels of compliance with guidelines might partly explain the reduced incidence of superficial SSIs among planned CDs; however, compliance had no effect on superficial SSI among emergency CDs or the incidence of deep SSIs. Conflict of interest The authors have no conflicts of interest. References [1] Smaill FM, Gyte GM. Antibiotic prophylaxis versus no prophylaxis for preventing infection after cesarean section. Cochrane Database Syst Rev 2010;1:CD007482. [2] Gall SA. The efficacy of prophylactic antibiotics in cesarean section. Am J Obstet Gynecol 1979;134:506–11. [3] Dillon WP, Seigel MS, Lele AS, O’Leary JA. Evaluation of cefoxitin prophylaxis for cesarean section. Int J Gynecol Obstet 1981;19(2):133–9.
5
[4] American College of Obstetricians and Gynecologists. ACOG practice bulleting number 47, October 2003: Prophylactic antibiotics in labor and delivery. Obstet Gynecol 2003;102(4):875–82. [5] van Schalkwyk J, Van Eyk N, Society of Obstetricians and Gynaecologists of Canada Infectious Diseases Committee. Antibiotic prophylaxis in obstetric procedures. J Obstet Gynaecol Can 2010;32(9):878–92. [6] National Institute of Public Health. Medical Birth Register. http://www.fhi.no/ helseregistre/medisinsk-fodselsregister. Published November 21, 2012. Accessed September 26, 2014. [7] Lichtenberg SM, Bungum L, Lande E. Cesarean section. In: Dalaker K, Berle EJ, editors. Clinical Guidelines in Obstetrics 1998 (in Norwegian). Oslo: The Norwegian Society of Obstetrics and Gynecology, The Norwegian Medical Association; 1998. [8] Häger R, Kolås T, Sviggum O, Novakovic Z. Cesarean Section. (in Norwegian) http:// legeforeningen.no/Fagmed/Norsk-gynekologisk-forening/Veiledere/veileder-ifodselshjelp-2008/. Published September 27, 2008. Accessed September 26, 2014. [9] Løwer HL, Eriksen HM, Aavitsland P, Skjeldestad FE. Methodology of the Norwegian Surveillance System for Healthcare-Associated Infections: the value of a mandatory system, automated data collection, and active postdischarge surveillance. Am J Infect Control 2013;41(7):591–6. [10] Hospital in Europe Link for Infection Control through Surveillance (HELICS), European Centre for Disease Prevention and Control. Surveillance of Surgical Site Infections: Master Protocol. http://www.ecdc.europa.eu/en/activities/surveillance/ HAI/Documents/0409_IPSE_SSI_protocol.pdf. Published September 2004. Accessed July 16, 2014. [11] Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1999;20(4):250–78. [12] Lovdata. Regulations on the Norwegian Surveillance System for antibiotic use and healthcare-associated infections (NOIS registry regulations). http://lovdata.no/ dokument/SF/forskrift/2005-06-17-611?q=Helseregister+forskriften. Published June 21, 2005. Accessed July 16, 2014. [13] Eriksen HM, Sæther AR, Økland I, Langen E, Sandness Y, Bødtker A, et al. Antibiotics prophylaxis in connection with caesarean section–guidelines at Norwegian maternity departments. Tidsskr Nor Laegeforen 2011;131(23):2355–8. [14] Doss AE, Davidson JD, Cliver SP, Wetta LA, Andrews WW, Tita AT. Antibiotic prophylaxis for cesarean delivery: survey of maternal-fetal medicine physicians in the U.S. J Matern Fetal Neonatal Med 2012;25(8):1264–6. [15] Committee opinion no. 465: antimicrobial prophylaxis for cesarean delivery: timing of administration. Obstet Gynecol 2010;116(3):791–2. [16] Olsen MA, Butler AM, Willers DM, Devkota P, Gross GA, Fraser VJ. Risk factors for surgical site infection after low transverse cesarean section. Infect Control Hosp Epidemiol 2008;29(6):477–86. [17] National Nosocomial Infections Surveillance System. National Nosocomial Infections Surveillance (NNIS) System Report, data summary from January 1992 through June 2004, issued October 2004. Am J Infect Control 2004;32(8):470–85. [18] Hofmeyr JG, Novikova N, Mathai M, Shah A. Techniques for cesarean section. Am J Obstet Gynecol 2009;201(5):431–44.
Please cite this article as: Skjeldestad FE, et al, The effect of antibiotic prophylaxis guidelines on surgical-site infections associated with cesarean delivery, Int J Gynecol Obstet (2014), http://dx.doi.org/10.1016/j.ijgo.2014.08.018