The use and abuse of antibiotics in elective colorectal surgery: The Saga Continues.

International Journal of Surgery (2005) 3, 69e74 www.int-journal-surgery.com REVIEW The use and abuse of antibiotics in elective colorectal surgery: The Saga Continues. Elie Aoun a,d, Sandra El Hachem a,d, Heitham Abdul-Baki a, Bassem Ayyach a, Mohamad Khalifeh b, Hani Chaar c, Zeina A. Kanafani a, Souha S. Kanj a, Ala I. Sharara a, * a Department of Internal Medicine, American University of Beirut Medical Center, Lebanon b Department of Surgery, American University of Beirut Medical Center, Lebanon c School of Pharmacy, Lebanese American University, Lebanon KEYWORDS Antimicrobials; Prophylaxis; Colectomy; Infection; Practice; Survey Abstract Background: The role of antibiotic prophylaxis in preventing postoperative complications in patients undergoing elective colorectal surgery is well established. Despite evidence that a single-dose prophylaxis is sufficient, the duration of antibiotic use in clinical practice is highly variable and surveys have identified persistent patterns of antibiotic abuse in elective colorectal surgery. Materials and methods: We conducted a retrospective review of all patients who underwent elective colorectal surgery between 1998 and 2002 at the American University of Beirut Medical Center. A survey among general surgeons in Lebanon was also performed to investigate the pattern of antibiotic prophylaxis used in such cases. The MEDLINE database (1966e2004) was searched for English-language articles and abstracts on antimicrobial use in elective colorectal surgery. Papers cited in relevant primary articles were also reviewed. Data were extracted and reviewed by all authors. Results: Two hundred and eleven matching patient-records were identified. A triple regimen including metronidazole, ampicillin and an aminoglycoside was the most commonly used preoperative prophylactic method. Patients received postoperative antibiotics for a mean of 6.66 G 2.62 days. The mean duration of postoperative antibiotic prophylaxis used by the interviewed surgeons was 4.31 G 1.08 days. Conclusions: Our study confirms that even when strong evidence exists, surgeons fail to adhere to antibiotic prophylaxis guidelines. This pattern is not unique to Lebanon but is shared to a large extent by surgeons around the world. Adherence to * Corresponding author. Box 16-B, American University of Beirut Medical Center, P.O. Box 11-0236, Riad El Solh 110 72020, Beirut, Lebanon. Tel.: C961 1 350000x5351; fax: C961 1 370814. E-mail address: as08@aub.edu.lb (A.I. Sharara). d These authors contributed equally to all aspects of the manuscript and both should be considered first authors. 1743-9191/$ - see front matter ª 2005 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijsu.2005.05.003

70 E. Aoun et al. published guidelines and improved education of surgeons are essential to the delivery of cost-effective medical practice. ª 2005 Surgical Associates Ltd. Published by Elsevier Ltd. All rights reserved. Introduction The use of antibiotic prophylaxis in patients undergoing elective colorectal surgery is well established. Controlled trials and a large meta-analysis have clearly shown a reduction in the incidence of post-operative wound infection in favour of antibiotic prophylaxis. 1e3 Although there is no clear consensus as to the optimal agent(s), the choice of antibiotics in elective colorectal surgery is guided by the fact that aerobic and anaerobic coverage is required. The duration of prophylaxis, however, is better defined. In 1998, a meta-analysis of existing randomized controlled trials demonstrated that single-dose antibiotic prophylaxis prior to elective colorectal surgery is as effective as a multiple-doses regimen and is therefore likely to result in reduced cost and less undesirable adverse events including the development of bacterial resistance and systemic toxicity. 3e9 Despite the level I evidence, general surgeons worldwide tend to administer antibiotic regimens for several days post-operatively in the absence of a valid medical indication. Hospital surveys in many countries have unveiled significant deficiencies in the appropriate use of antibiotic prophylaxis and a lack of adherence with existing guidelines in colon surgery. 10e15 In this study, we aimed to examine the practice of general surgeons in Lebanon regarding antibiotic prophylaxis in elective colorectal surgery. For that purpose, a retrospective review of elective colorectal surgeries in a University hospital was conducted followed by a survey of Lebanese general surgeons at large. A systematic review of the medical literature was done to compare existing practices in various countries as relates to antibiotic prophylaxis in elective colorectal surgery. Materials and methods Patients The first component of the study consisted of a retrospective analysis of medical records of all patients who underwent elective colorectal surgery between 1998 and 2002 at the American University of Beirut Medical Center, a tertiary care center in Lebanon. Relevant data were collected from the charts using a specific form that included patients general characteristics and details about the operation itself including the indication for surgery, the technique used, the wound class, and the duration of the procedure. The use, type and duration of antibiotics were collected along with outcomes such as the occurrence of post-operative fever, wound infections, space infections, anastomotic leaks, and extra-abdominal complications including pneumonia, urinary tract infections and phlebitis. Survey We conducted a cross-sectional survey using a structured questionnaire administered to general surgeons practicing in several hospitals throughout Lebanon. The questionnaire was designed to seek information on the surgeon s practice, level of experience, and preferences regarding the type, timing and duration of antibiotics used in the prophylaxis in elective colorectal surgeries. Literature review The MEDLINE database (1966e2004) was searched for English-language articles and abstracts on antimicrobial prophylaxis in elective colorectal surgery. Papers cited in relevant primary articles were also reviewed. Original and review articles were evaluated, and the most relevant were selected. Data were extracted and reviewed by all authors. Statistical analysis Statistical analysis was performed using SPSS version 11.5. Cross-tabulation, frequency tables and c 2 tests were derived in order to detect any associations between the different variables under study. Results Patients characteristics A total of 211 matching patient-records were identified. Table 1 lists the baseline characteristics

Antibiotics in elective colon surgery 71 of the patients: 127 patients (60.2%) were males and 84 (39.8%) were females. The mean age was 60.53 G 13.06 years. Concurrent medical conditions at the time of surgery included diabetes (16.1%), coronary artery disease (13.7%), chronic obstructive pulmonary disease (2.8%), hypertension (16.1%) and chronic liver disease (1.9%). Fortysix patients (21.8%) had previously undergone abdominal surgery. Sixteen patients (7.6%) were on immunosuppressive medications at the time of the operation. The most common indications for surgery were cancer (75.4%), inflammatory bowel disease (6.2%) and diverticular disease (5.7%). The mean preoperative hemoglobin level was 11.70 G 1.89 mg/dl. Bowel preparation was done using polyethylene glycol electrolyte solution (PEG-E) in 117 cases (55.5%), combination of enema and PEG-E in 52 cases (24.6%). The remaining patients were prepared using either enema alone or oral antibiotics. Surgery An open surgery was performed in 180 (85.3%) cases, and laparoscopically in the remaining 31 (14.7%) patients. The most common procedure performed was a hemicolectomy (51.7%) followed by total colectomy (23.7%), low anterior resection (17.5%) and sigmoidectomy (4.7%). The mean procedure duration was 216.84 G 74.27 min. The wound was classified as clean-contaminated in 134 cases (63.5%) and contaminated in 68 (32.2%) cases. The remaining 9 cases were classified as having dirty wounds. A drain was used in 82.9% of cases. Blood transfusions (between 1 and 3 units) were used in 74 cases (35.1%) during the operation. Table 1 General characteristics of the 211 cases reviewed Gender N (%) Male 127 (60.2) Female 84 (39.8) Mean age (G SD) 60.53 G 13.06 years Diabetes N (%) 34 (16.1) Hypertension N (%) 34 (16.1) Coronary artery 29 (13.7) disease N (%) Previous cerebrovascular 22 (10.4) accident N (%) Chronic obstructive 6 (2.8) pulmonary disease N (%) Chronic liver disease N (%) 4 (1.9) Previous abdominal 46 (21.8) surgery N (%) Immunosuppression N (%) 16 (7.6) Preoperative antibiotics use All patients received preoperative antibiotic prophylaxis. The most commonly used regimen were triple therapy including metronidazole, ampicillin and an aminoglycoside (44.7%) and piperacilline tazobactam (43.7%). Sporadic regimens including quinolones, macrolides and cephalosporins were used in the remaining cases. Most of the patients received one dose of prophylaxis on call to the operating room (84.5%) while 14.1% of patients received three doses of antibiotics preoperatively. Post-operative antibiotic use All patients received post-operative antibiotics, most commonly triple therapy (metronidazole, ampicillin and an aminoglycoside) in 42.2% of cases and piperacillinetazobactam in 33.6%. These regimen were used for an average of 6.66 G 2.62 days post-operatively. In addition, 18 patients (8.5%) were discharged home on antibiotics for one week. Post-operative complications Thirty-six patients (17.1%) developed post-operative complications: 13 patients (6.2%) developed pneumonia, 3 (1.4%) a urinary tract infection, 19 (9%) a superficial wound infection, and deep wound and compartment infection in 7 (3.3%) of the patients (Table 2). All 36 developed fever postoperatively and wound cultures were taken in all these cases and were positive in 23 patients (74.2%) growing Escherichia coli (47.8%), Klebsiella species (21.7%) and both E. coli and Enterococci (21.7%). Of the 36 patients, 30 (83.3%) were shifted to a new antibiotic regimen, most commonly carbapenem (50%), vancomycin and quinolones (10%), and teicoplanin (6.7%). Abdominal and pelvic abscesses were diagnosed in one patient. Staphylococci grew in the culture from the abdominal abscess and E. coli in the culture from pelvic abscess. None of the patients Table 2 Post-operative complications Complication n % Fever 36 17.1 Pneumonia 13 6.2 Urinary tract infection 3 1.4 Superficial wound infection 19 9.0 Deep wound infection 7 3.3 Abdominal abscess 1 0.5 Pelvic abscess 1 0.5 No. of patients with complications 36 17.1

72 E. Aoun et al. developed an anastomotic leak. The average postoperative duration of hospital stay was 9.83 G 3.31 days. Upon cross-tabulation of the data, there was no statistical correlation between the duration of post-operative antibiotic and the development of post-operative complications. Survey results A total of 51 self-administered questionnaires were filled and returned by general surgeons practicing in several hospitals in Lebanon. Table 3 shows the results of the survey. The average duration in general surgery practice was 11.61 G 3.45 years, and the mean number of elective colorectal surgeries per year was 19.92 G 6.77. Most interviewed surgeons (88.2%) indicated that laboratory data are an important criterion in the choice of preoperative regimen chosen while 31 (60.8%) considered the presence of co-morbid conditions as an important factor in making such a choice. The most commonly used regimens were triple therapy (ampicillin, metronidazole and an aminoglycoside) (49%) and a regimen including a cephalosporin and metronidazole (43.1%). Thirtyeight surgeons (74.5%) stated that they use the same regimen post-operatively. The mean duration of post-operative prophylactic antibiotics used was 4.31 G 1.08 days; 27.5% considered that patients undergoing elective colorectal surgery should receive a longer course of antibiotic prophylaxis (i.e. O5 days) on an in-patient basis and 49% indicated that they usually discharge their patients on oral antibiotics for one week. There was no significant difference amongst universitybased and community-based general surgeons regarding the use and duration of post-operative Table 3 Survey results Number of surgeons surveyed 51 Years of surgery practice (mean G SD) 11.61 G 6.81 Number of elective colectomies/year 19.92 G 9.77 (mean G SD) Choice of preoperative antibiotics (%) Triple regimen a (49.0) CPH C metronidazole (43.1) Piperacillinetazobactam (7.9) Days of antibiotic use post-operative (mean G SD) Use of oral antibiotics on discharge (%) 49% 4.31 G 1.086 a Triple regimen: ampicillin, metronidazole and an aminoglycoside; CPH: cephalosporin. antibiotics. Further analysis also showed no correlation between the experience of the surgeon (as defined by years of experience and number of yearly performed surgeries) and the duration of post-operative antibiotic. When asked to indicate which factors predispose patients to post-operative complications, 88.2% considered intra-operative complications and 56.9% indicated surgical expertise as risk factors. The patient s age and the presence of co-morbid conditions were not considered as risk factors. The surveyed physicians indicated that their choice regarding the duration of post-operative antibiotics is affected by an inadvertent enterotomy (86.3%) and by a history of previous laparotomy (78.4%). The majority did not consider the area of the colon operated as a factor in the duration of post-operative antibiotics. Discussion Mortality from colorectal surgery used to exceed 20% before the advent of new surgical techniques, mechanical colon cleansing, and better patient care including the use of antibiotic prophylaxis. 1 Sepsis was the most common cause of death after surgery. Despite the marked decrease in mortality rates after colorectal surgery, infectious complications result in significant morbidity leading to increased costs and hospital stay. 9 The most common infectious complications occurring after colorectal surgery include surgical wound infection, abdominal and pelvic abscesses, and the development of anastomotic leaks. The main organisms involved in the pathophysiology of these complications are the colonic flora, mostly anaerobic bacteria including Bacteroides and Clostridia species, and aerobic organisms (E. coli, Proteus, Klebsiella and Pseudomonas species). 15 The use of antibiotic prophylaxis has been shown to significantly reduce the incidence of such complications. Indeed, almost 40% of patients would develop wound infection after colorectal surgery if deprived from prophylaxis. The risk drops to 11e22% with the use of peri-operative antibiotics. 3e9 Controlled trials and a large meta-analysis have demonstrated that single-dose antibiotic prophylaxis is as effective as multiple-dose regimens in elective colorectal surgery. Despite the evidence, antibiotic prophylaxis is usually extended in the post-operative period. Our literature review identified 93 relevant articles: Table 4 lists the published studies that have examined the pattern and practice of antibiotic prophylaxis in elective colorectal surgery. The majority of these studies involved cross-sectional

Antibiotics in elective colon surgery 73 surveys of surgeons and a few have examined the actual existing practice using a retrospective review of medical records. Despite inherent inaccuracies, these studies show that 21e99% of surgeons do not adhere to the existing evidence on antibiotic prophylaxis in elective colorectal surgery. 10e15 Our study confirms that this incorrect practice pattern is also highly prevalent in Lebanon amongst university and community surgeons alike and that surgical expertise and experience did not affect the duration of antibiotic use. The reasons for lack of adherence to existing evidence and guidelines are many and vary widely from ignorance of evidence-based practice, inadvertent neglect, reliance on habit, conformity to peers or senior surgeons (in the case on interns or residents), to fear of litigation. The perception that the extension of antibiotic prophylaxis to the post-operative period may offer the surgeon added comfort without prohibitive cost or undue risk to the patient is incorrect and gravely underestimates the collateral damage and the hidden costs. These include prolonged hospital stay, drug-related adverse events, Clostridium difficile colitis and most importantly the emergence of antibiotic resistance. In fact, our study shows that the extension of antibiotic prophylaxis post-operatively results in an incidence of post-operative infectious complications that is not dissimilar from published studies using single-dose prophylaxis (17.1%). Although lacking in our analysis, the direct and indirect costs of prolonged antibiotic use cannot be overestimated. Our study confirms that even when strong evidence exists, surgeons in Lebanon fail to adhere to guidelines. As shown in Table 4, this pattern is not unique to Lebanon but is shared almost universally by surgeons from various countries including in Europe and North America. It is improbable that surgeons have no access to the scientific facts given the current growth in evidence-based medicine evident in clinical practice guidelines and consensus conferences. Our study and that of others 12,16e19 suggest that the root of Table 4 Comparison of studies of prophylactic antibiotic use in elective colorectal surgery Author Country Year Type of study Conclusion Comments McDonald and Karran 13 Mercer et al. 14 United Kingdom New Zealand 1982 Survey (32 surgeons) 1991 Survey (118 surgeons) Kappstein and Germany 1991 Survey Daschner 12 (889 surgical departments) Widdison et al 17 United Kingdom 1993 Survey (160 surgeons) Gul et al. 10 Malaysia 2002 Survey (96 surgeons) Wasey et al. 11 Canada 2003 Retrospective chart review (103 patients) Zmora et al. 15 USA 2003 Survey (515 surgeons) Aoun et al. (current study) Lebanon 2004 Retrospective chart review (211 patients) and survey (51 surgeons) 50% used peri-operative antibiotics for!2 days 21.2% used peri-operative antibiotics only Duration of prophylaxis correct in 42.9% of colorectal surgery cases 79% recommended prophylaxis for!24 h 4/118 did not use prophylactic antibiotics Duration of prophylaxis incorrect in the vast majority of surgical practices (general, orthopaedic, trauma, and cardiothoracic) Colorectal, prosthetic joint and gastroesophageal surgeries combined 72% used more than one dose No difference between academic and private practice 95% incorrect use (O1 dose) 85% continued Rx post-operative for an average of 2 doses (range 1e15) 99% incorrect use (O1 dose). None of the surveyed surgeons uses peri-operative antibiotics for %1 day Only study examining both actual clinical practice (retrospective review) and surgeons preferences (survey)

74 E. Aoun et al. the problem may be a deficiency in training and knowledge of evidence-based guidelines in the surgical disciplines. The promotion of cost-effective medical practice remains a key demand of patients, physicians, governments and insurance carriers. It is the duty of surgeons, hospitals infection control and quality assurance committees as well as surgery training programs and governing bodies to assume an added responsibility in ensuring strict adherence to correct education and practice of antibiotic prophylaxis in elective colorectal surgery. References 1. Baum ML, Anish DS, Chalmers TC, Sacks HS, Smith Jr H, Fagerstrom RM. A survey of clinical trials of antibiotic prophylaxis in colon surgery: evidence against further use of no-treatment controls. N Engl J Med 1981;305:795e9. 2. Bartlett SP, Burton RC. Effects of prophylactic antibiotics on wound infection after elective colon and rectal surgery: 1960 to 1980. Am J Surg 1983;145:300e9. 3. Song F, Glenny AM. Antimicrobial prophylaxis in colorectal surgery: a systematic review of randomized controlled trials. Br J Surg 1998;85:1232e41. 4. Juul P, Klaaborg KE, Kronborg O. Single or multiple doses of metronidazole and ampicillin in elective colorectal surgery. A randomized trial. Dis Colon Rectum 1987;30:526e8. 5. Aberg C, Thore M. Single versus triple dose antimicrobial prophylaxis in elective abdominal surgery and the impact on bacterial ecology. J Hosp Infect 1991;18:149e54. 6. Carr ND, Hobbiss J, Cade D, Schofield PF. Metronidazole in the prevention of wound sepsis after elective colorectal surgery. J R Coll Surg Edinb 1984;29:139e42. 7. Kow L, Toouli J, Brookman J, McDonald PJ. Comparison of cefotaxime plus metronidazole versus cefoxitin for prevention of wound infection after abdominal surgery. World J Surg 1995;19:680e6 [discussion 686]. 8. Cuthbertson AM, McLeish AR, Penfold JC, et al. A comparison between single and double dose intravenous Timentin for the prophylaxis of wound infection in elective colorectal surgery. Dis Colon Rectum 1991;34:151e5. 9. Brachman PS, Dan BB, Haley RW, Hooton TM, Garner JS, Allen JR. Nosocomial surgical infections: incidence and cost. Surg Clin North Am 1980;60:15e25. 10. Gul YA, Lian LH, Jabar FM, Moissinac K. Antibiotic prophylaxis in elective colorectal surgery. ANZ J Surg 2002;72:275e8. 11. Wasey N, Baughan J, de Gara CJ. Prophylaxis in elective colorectal surgery: the cost of ignoring the evidence. Can J Surg 2003;46:279e84. 12. Kappstein I, Daschner FD. Use of perioperative antibiotic prophylaxis in selected surgical procedures e results of a survey in 889 surgical departments in German hospitals. Infection 1991;19:391e4. 13. McDonald PJ, Karran SJ. Preoperative antimicrobial prescribing practice for elective colorectal surgery in Wessex. Lancet 1981;1982(2):753e5. 14. Mercer PM, Bagshaw PF, Utley RJ. A survey of antimicrobial prophylaxis in elective colorectal surgery in New Zealand. Aust N Z J Surg 1991;61:29e33. 15. Zmora O, Wexner SD, Hajjar L, Park T, Efron JE, Nogueras JJ, et al. Trends in preparation for colorectal surgery: survey of the members of the American Society of Colon and Rectal Surgeons. Am Surg 2003;69:150e4. 16. Janknegt R, Wijnands WJ, Stobberingh E. Antimicrobial prophylaxis in bowel surgery in The Netherlands. Eur J Clin Microbiol Infect Dis 1994;13:596e600. 17. Widdison AL, Pope NR, Brown EM. Survey of guidelines for antimicrobial prophylaxis in surgery. J Hosp Infect 1993;25: 199e205. 18. Delgadillo J, Ramirez R, Cebrecos J, Arnau JM, Laporte JR. [The use of antibiotics in surgical prophylaxis. The characteristics and consequences]. Med Clin (Barc) 1993; 100:404e6. 19. Lim VK, Cheong YM, Suleiman AB. The use of surgical antibiotic prophylaxis in seven Malaysian hospitals. Southeast Asian J Trop Med Public Health 1994;25:698e701.