Current Practices of Preoperative Bowel Preparation Among North American Colorectal Surgeons

Transcription

1 609 Current Practices of Preoperative Bowel Preparation Among North American Colorectal Surgeons Ronald Lee Nichols, Jeffrey W. Smith, Rena Y. Garcia, From the Department of Surgery, Tulane University School of Medicine, Ruth S. Waterman, and James W. C. Holmes New Orleans, Louisiana In North America, the rate of infections following colorectal surgery decreased after the introduction of oral antibiotic bowel preparation against colonic microflora. Eight hundred eight boardcertified colorectal surgeons were surveyed for their current bowel preparation practices before elective procedures. The 471 responders (58%) all use mechanical preparation: oral polyethylene glycol solution (70.9% of the respondents), oral sodium phosphate solution with or without bisacodyl (28.4%), and "traditional" methods of dietary restriction, cathartics, and enemas (28.4%). Most surgeons (86.5%) add oral and parenteral antibiotics to the regimen; 11.5% add only parenteral antibiotics, 1.1% add only oral antibiotics, and 0.9% add no antibiotics. Generally (77.8% of cases), oral neomycin and erythromycin or metronidazole are combined with a perioperative parenteral antibiotic. Most individuals start the preparation as outpatients the day before surgery, and the parenteral drugs are added to the regimen 1 2 hours before the procedure. The use of outpatient bowel preparation is increasing; however, patient selection is critical, and education is needed to reduce the rate of complications. The true role of colonic intraluminal bacteria, both facultative and anaerobic, in the etiology of infectious complications following colorectal surgery was clarified 25 years ago [1-4]. Both the colonic bacterial burden and the rate of subsequent infections were significantly decreased when the preoperative bowel preparation included orally administered antibiotics effective against both bacterial types [3, 4]. Specifically, it was shown that mechanical preparation and a three-dose oral antibiotic regimen consisting of 1 g each of erythromycin base and neomycin resulted in suppression of the facultative and anaerobic constituents of the colonic and fecal microflora. Currently, it is generally accepted that effective bowel preparation includes various oral or parenteral antibiotics, alone or in combination, that have aerobic and anaerobic activities combined with an effective mechanical preparation [5]. Many different antibiotic regimens have been proposed and tested clinically, with some yielding better results than others. Although originally only oral antibiotics were used effectively, in current practice, they are now most often combined with perioperative parenteral antibiotics. Various mechanical preparations have also been used to reduce the gross intraluminal contents during the surgical procedures. A previous survey done in 1988, and reported in 1990 [6], showed that the most preferred bowel preparation at that time was oral polyethylene glycol (PEG) solution for mechanical Received 30 July 1996; revised 1 October This work was presented in part at the 9th Annual Meeting of the Surgical Infection Society-Europe held on 30 May to 1 June 1996 in Paris. Reprints or correspondence: Dr. Ronald Lee Nichols, Department of Surgery (SL-22), Tulane University School of Medicine, 1430 Tulane Avenue, New Orleans, Louisiana Clinical Infectious Diseases 1997; 24: by The University of Chicago. All rights reserved /97/ $02.00 cleansing combined with preoperative oral neomycin/erythromycin base and a perioperative parenteral second-generation cephalosporin antibiotic. Since the time frame of the previous survey, several new antibiotics have become available for use, older agents have become generic and their prices have been reduced, and additional clinical studies of various bowel preparations have been conducted [5]. There has also been an increased influence of managed care approaches to treatment in the interests of cost containment. An impetus toward preoperative bowel preparation to be conducted on an outpatient basis, commonly at the patient's home, has likewise gained support [7]. There remains some controversy over which antibiotics provide the optimal prophylaxis; the duration of preparation; whether oral, parenteral, or a combination is preferred; and which mechanical method should be used. In an attempt to gather current knowledge of North American bowel preparation practices before elective colorectal procedures, we sent a survey to all currently active board-certified colorectal surgeons in the United States (including Puerto Rico) and Canada (see appendix at the end of the text). Methods The names and addresses of all currently active board-certified colon and rectal surgeons in the United States and Canada were obtained from the American Society of Colon and Rectal Surgeons (Arlington Heights, IL). These physicians were sent a questionnaire inquiring about their preoperative bowel preparations before elective surgical procedures. The 20 questions covered demographics, patient numbers and types, and both mechanical and antibiotic preparative techniques. Specific questions concerned the use of oral vs. parenteral antibiotics, preferred mechanical cleansing procedures, and the total duration of the preparation.

2 610 Nichols et al. CID 1997;24 (April) In many questions, the responses added up to more than the number of surgeons responding because the questionnaires asked open-ended questions rather than limiting or forcing choices to one answer. A Likert scale (a five value rating scale ranging from "not important" to "very important" ) was used to evaluate the factors influencing the overall choice of bowel preparations. Eight hundred eight questionnaires were sent in two mailings in March and May The respondents were asked to return the anonymous questionnaires by either prepaid business reply mail or toll-free facsimile. The responses were analyzed by use of Statview 4.01 (Abacus Concepts, Berkeley, CA) on a Macintosh PowerBook Results Participating surgeons. Four hundred seventy-one (58%) of the 808 colon and rectal surgeons who were sent questionnaires returned their surveys within 4 months, and their responses were analyzed for this report. Responses were received from 45 states, the District of Columbia, Puerto Rico, and three Canadian provinces. Response rates for the states or provinces with more than five board-certified surgeons ranged from 44% to 88%. Most surgeons (391 [83%]) identified their community size as large (population, >100,000); 72 (15.3%), as medium (population, 25, ,000); and only 8 (1.7%), as small (population, <25,000). The most commonly reported medical affiliations were a community hospital (62.6%), teaching hospital (44.4%), or a large medical center (33.3%) (table 1). The responding surgeons received their colon and rectal surgery board certification an average of 11.4 years ago (range, 1-39 years). Only three surgeons had been recertified in colon and rectal surgery, all since Four hundred forty-seven surgeons indicated the number of procedures that they perform each month. They reported approximately equal average numbers of colon (6.1) and rectal (8.3) procedures each month and twice as many anal procedures Table 1. Location of professional practices of respondents to a survey on North American bowel preparation practices before elective colorectal procedures. Practice type No. of affiliations* Percent of respondents Community hospital Teaching hospital Large medical center Medical school Veterans hospital Military hospital Health maintenance organization * Respondents reported all affiliations; the total number of affiliations is greater than the number (471) of surgeons responding. (15.4) (table 2). Most of their procedures are elective (86%-91% depending upon the type of procedure), with most of the patients being admitted to the hospital on the day of surgery (65%-81%) following completion of the bowel preparation at home. Mechanical procedures. All 471 surgeons who reported their bowel preparative procedures routinely use some form of mechanical preparation with their patients (figure 1). The most commonly preferred forms of mechanical bowel preparation are oral PEG solution (70.9% of respondents), oral sodium phosphate-buffered solution with or without bisacodyl (28.4%), or the "traditional" usage of dietary restriction, cathartics (including magnesium citrate or sulfate), and enemas (28.4%) (table 3). Only a small number of surgeons reported that they routinely use whole-gut irrigation, mannitol, or other methods. The preferred time to start the mechanical preparation is usually hours before the surgical procedure (figure 2). Although there is some variation in the timing, all respondents start the preparation r 24 hours before the procedure. Most patients complete this mechanical preparation on an outpatient basis at home before hospital admission. The traditional bowel preparation, when used, is started an average of 29.8 hours (range, hours) before the surgical procedure. Those surgeons preferring PEG solution employ an average of 3.7 L (range, 1-8 L) over 3-4 hours (range, 1-24 hours). Although the participants reported their most commonly used mechanical methods, they would consider other procedures when they thought that it was in the patients' best interests or when it was medically indicated. Factors in the decisionmaking process include noncooperative or noncompliant patients; those who are very young, old, or frail; or those with disease states that might be compromised (e.g., severe diverticulitis; active colitis; inflammatory bowel disease; pulmonary, cardiac, or renal disease; severe nausea, cramping, or constipation; partial obstruction; or tight strictures). Antibiotic prophylaxis. Of 468 respondents, almost all (464 [99.1%]) reported that they routinely use preoperative prophylactic antibiotics (table 4). Most (391 [85.4%]) of the 458 surgeons who listed the rationale for antibiotic choice reported that the antibiotics should protect against facultative and anaerobic colonic bacteria. Smaller numbers of surgeons were concerned with either aerobes alone (42 [9.2%]) or anaerobes alone (25 [5.4%]). Correspondingly, 45.9% (194) of 423 surgeons reported that both facultative and anaerobic bacteria were responsible for infections following colorectal procedures at their hospitals. Problems with facultative bacteria alone were reported by 43.7% (185) of the surgeons, while only 8% (34) related concerns solely with anaerobic infections. Ten surgeons (2.4%) were unsure of the bacterial cause of infections at their institutions. The choices of oral and parenteral antibiotics are listed in table 5. The 471 surgeons who responded reported a total of 711 different antibiotic regimens, 625 of which include oral antibiotics. The many regimens had slight variations in the antibiotic and dosage choices, and for this report, the parenteral

3 CID 1997;24 (April) Preoperative Preparation of the Colon 611 Table 2. Operative procedures performed per month by 447 respondents to a survey on North American bowel preparation practices before elective colorectal procedures. Average no. Percent of patients Type of of procedures Percent of Percent of admitted on the procedure per month (range) emergent procedures elective procedures day of surgery Colon 8.3 (1-60) Rectal 6.1 (0-60) Anal 15.4 (1-200) choices were combined into antibiotic families. As can be seen, the most prevalent oral regimens are neomycin with either erythromycin or metronidazole. In most cases (77.8% of the 711 regimens), oral neomycin and either erythromycin or metronidazole are combined with the perioperative use of a parenteral antibiotic. The oral antibiotics are normally started on the day before surgery (97.2%), with only 11 respondents (2.8%) beginning them earlier than 1 day before (figure 2). Of all reported regimens, the parenteral antibiotics most often chosen are second-generation cephalosporins (figure 1). Some surgeons prefer to use a firstor third-generation cephalosporin, penicillin with a /3-lactamase inhibitor, or intravenous metronidazole. Most surgeons start the parenteral antibiotics 1 hour before surgery (figure 2). Most (93.7%) of the surgeons limit the routine administration of the parenteral antibiotics to four or less doses stopping within 24 hours after surgery. A few, however, do give the drugs for 2 to 4 days. The surgeons stated that the antibiotic regimens would be continued if perforation or spillage was noted during surgery. They also would consider alternate regimens, presumably increased duration or different drugs, for immunocompromised patients or those with Crohn's disease, prosthetic devices, antibiotic allergies, or cardiac valve replacements. Influencing factors. Six factors were addressed concerning the surgeons' choices for bowel preparation. The results showed that the most important concerns are reduced rates of infections in their patients that result from both reduction of bacterial burden and a grossly clean colon at the time of operation (table 6). Patient acceptability and ease of administration were of lesser importance, while the least important concern was the cost of the preparation. Specific questions concerning the use of a home bowel preparation before surgery were also addressed. Although most surgeons (283 [59.5%] of 476) thought it was as good as inpatient administration, approximately one-third (146 [30.7%]) did not agree, and 10% (47) thought it was usually all right but did express concerns for some patients. The concerns noted by the surgeons who responded "no" and "usually OK" included such problems as dehydration, lack of compliance, the patients' inability to self-administer an effective enema, and an inadequately prepared colon found during surgery. They believed that certain patients should not receive home bowel preparations unless there was adequate supervision by a family member or visiting nurse. These patients would be elderly individuals, those with disabilities, nonreliant or nonmotivated patients, or those distracted or overly anxious over the impending surgery. Only a small number Figure 1. Preoperative bowel preparation regimens currently prescribed in North America. All 471 surgeons answering the survey reported the use of mechanical preparation. Most (70.9%) of the surgeons use oral polyethylene glycol solution, but equal use of oral sodium phosphate solution (28.4%) or "traditional" preparative techniques (cathartics and enemas; 28.4%) was also reported. Antibiotics are added to the regimen by 99.1% of the surgeons, with most employing both oral and parenteral types. * = percent of survey respondents; ** = percent of 711 antibiotic regimens reported. 70.9%* Polyethylene glycol solution (po) 0.9%* None Oral only I I Liquid or low-residue diet I Mechanical preparation 28.4%* Sodium phosphate solution (po) Antibiotic preparation Oral components Neomycin plus erythromycin 53.4%** plus metronidazole 35.0%*" Oral plus 86.5%* parenteral 28.4%* "Traditional" cathartics and enemas Parenteral components 11.5%* Parenteral only Second-generation cephalosporin 65.4%** cephalosporin 6.8%** Penicillin/inhibitor combination 8.3%** Metronidazole 7.5%**

4 612 Nichols et al. CID 1997; 24 (April) Table 3. Mechanical preparations used by respondents to a survey on North American bowel preparation practices before elective colorect4l procedures. Mechanical preparation used No.* of preparations Percent of respondents Polyethylene glycol solution Sodium phosphate solutions "Traditional" Whole-gut irrigation Mannitol successful in suppressing intraluminal bacteria when administered in 1-g doses at 1 P.M., 2 P.M., and 11 P.M. on the day before the surgical procedure [3, 4]. Pharmacokinetic studies showed that neomycin is not absorbed and remains bacteriologically active within the lumen of the colon, while high intraluminal and serum levels of erythromycin are found at the time of surgery (8 A.M.) [ 12, 13]. Both intraluminal (local) and serum (systemic) antibiotics are thought to contribute toward reducing the occurrence of postoperative infections [5]. * Respondents reported their normally used preparation(s); the total number (626) of preparations is more than the number (471) of surgeons responding. All surgeons used some form of mechanical preparation. t With or without bisacodyl. I Combination of dietary restriction, enemas, and cathartics. of surgeons thought that a home preparation should only be performed for colonoscopy and that an inpatient preparation should be used for all other procedures. Discussion During colon and rectal surgical procedures, it is important to avoid bacterial contamination of the peritoneal cavity or adjacent tissues by colonic microflora to prevent serious postoperative intraabdominal or surgical site infections. For most of this century, surgeons have tried to sterilize the lumen of the colon to reduce the rates of surgical morbidity and mortality following colon or rectal surgery [2]. As early as 1951, Finegold [8] reported on the effects of various antimicrobials on the colonic microflora. Although coliform bacteria were suppressed, the anaerobes were not significantly affected. Before the 1970s, the primary method of reducing the bacterial burden was through effective mechanical cleansing [2]. In 1971, studies of traditional mechanical cleansing (dietary restrictions, cathartics, and enemas) showed that although gross lumps of stool were removed, bacterial counts in the remaining liquid colonic contents were still significant [9]. Oral antibiotics used at that time (e.g., sulfonamides, streptomycin, kanamycin, and neomycin) had activities effective in suppressing facultative bacteria alone but often failed to prevent postoperative infection [2, 3, 10]. In addition, the oral antibiotics were given for up to 5 days, resulting in intracolonic overgrowth of staphylococci or yeast. Consequently, many' believed that antibiotic prophylaxis was of little use and did not routinely employ it [2]. In the early 1970s, it was found that the intraluminal anaerobic microflora of the colon and rectum greatly outnumbered the facultative organisms ( 1,000 to 1) [10, 11]. In , it was shown that the addition of an antibiotic effective against the predominant anaerobic bacteria (oral erythromycin base) to an antibiotic previously shown to be effective against facultative organisms (oral neomycin) was Figure 2. Timing of the mechanical and antibiotic components of the preoperative bowel preparation in North America. Most surgeons start the mechanical procedures within hours before the operative procedure. The oral antibiotics are administered hours before, with parenteral antibiotics being added to the regimen within 1 hour of the procedure.

5 CID 1997; 24 (April) Preoperative Preparation of the Colon 613 Table 4. Use of oral and parenteral antibiotic bowel preparative techniques by respondents to a survey on North American bowel preparation practices before elective colorectal procedures. Use of oral antibiotics No. (%) using parenteral antibiotics Yes No Yes 405 (86.5) 5 (1.1) No 54 (11.5) 4 (0.9) NOTE. Four hundred sixty-eight of 471 surgeons responded to this question; all reported the use of mechanical bowel preparation in their preoperative regimen. Although the early reports showed the efficacy of oral prophylaxis in suppressing the colonic microflora, later studies tested the idea that parenteral antibiotics added to or substituted for the oral agents could also be effective [5]. Many different regimens comparing a multitude of antibiotics were studied with varying and conflicting results [5]. On the basis of these reports, some surgeons, predominantly those in Europe, prefer systemic parenteral agents alone, whereas North American surgeons favor a combination of oral and parenteral agents [14]. It is important that one be cautious when evaluating the results of prophylactic studies. For example, results of a twocenter trial that were published independently showed striking differences in the rates of infections between the two arms: oral neomycin/erythromycin and parenteral metronidazole/ceftriaxone (site 1, 41% and 10%, respectively; site 2, 4% and 7%, respectively) [15, 16]. A questionable study design was later noted, since mechanical cleansing was used only at the second hospital [17]. Therefore, it is imperative that multicenter trials be published together to enable readers to make informed judgments based upon all available data. Previous surveys have shown that the percentage of North American colon and rectal surgeons using effective antibiotic prophylaxis has increased from 85% in 1979 to 100% in 1988 [6, 18, 19]. In 1976, one survey indicated that 6% of surgeons did not use antibiotic prophylaxis but relied upon mechanical techniques to reduce the rate of postoperative infections [20]. It was also seen that systemic antibiotics alone were used by 8% of the respondents, oral antibiotics alone were used by 37%, and a combination of both were used by 49%. A sizable percentage (18%) of the surgeons began to administer the systemic antibiotics postoperatively, a practice now known to be suboptimal. Although a small percentage (0.9%) of surgeons still fail to use effective antibiotic prophylaxis, our survey indicates that it remains the standard of care in North America. In 1990, a comparison with British surgeons indicated that 92% used antibiotic prophylaxis routinely [21]. However, only 17% used topical (oral) antibiotic prophylaxis. Seventy-eight percent of the British surgeons favored a regimen of parenteral cephalosporin plus metronidazole. A three-dose regimen (one during surgery and two postoperatively) was reported by 43% of the surgeons, while an additional 48% continued to administer the antibiotics beyond the three doses. Although the antibiotic combination should be effective as preoperative prophylaxis, both the duration of administration and the use of mechanical preparations with enemas (3 8%), purgatives (37%), mannitol (19%), or whole-gut irrigation (6%) are in sharp contrast to North American practices and might help explain the traditionally higher rates of wound infection in the United Kingdom [5]. In the United States, rates of postoperative infection following administration of appropriate oral agents, with or without the addition of a parenteral agent, have been constantly reported to be < 10% among patients without additional risk factors for infection [22]. Table 5. Most commonly used oral and parenteral antibiotics for preoperative bowel preparation before elective colon or rectal surgery in a North American survey. No. using parenteral antibiotic(s) Cephalosporin Oral antibiotic(s) First-generation Second-generation Third-generation 0-Lactamase inhibitor combinations Metronidazole Total Neomycin Plus clindamycin Plus erythromycin Plus metronidazole Plus erythromycin and metronidazole Metronidazole Plus erythromycin Total with oral antibiotics Total without oral antibiotics Total NOTE. Four hundred seventy-one respondents listed all regimens commonly prescribed by them.

6 614 Nichols et al. CID 1997;24 (April) Table 6. Factors influencing choices of bowel preparation in a survey on North American bowel preparation practices before elective colorectal procedures. Factor No. of respondents Mean score* Mode Median Reduced rate of infections Reduction of bacteria at operative site Cleanliness of operative site Patient acceptability Ease of administration Cost * Higher scores indicate greater importance. Choices varied from 1 (not important) to 5 (very important). Our survey of clinically active colorectal surgeons found that oral antibiotics remain well accepted and that the rate of use (91.8%) is similar to that reported in 1990 (87.6%) [6]. A more limited survey from the same period showed that 87% of 206 surgeons used oral prophylaxis [19], while one conducted in 1976 indicated its use by 86% of 582 surgeons [20]. From 1988 to the present, the rate of the use of parenteral antibiotics, with or without oral agents, has slightly increased from 96.6% to 98.1% [6], which is in contrast to the rates reported for 1988 and 1976 (90% and 57%, respectively) [19, 20]. Parenteral antibiotics are usually administered within 1 hour of surgery, a time frame that will provide adequate serum and tissue levels at the time of the procedure. The practice of having "on call" parenteral agents in the operating room should be discouraged, as it often results in inadequate tissue levels during the procedure. In 1988, 63% of surgeons continued to administer parenteral antibiotics 1 day postoperatively, and 25% continued their use for 2-3 days postoperatively [19]. Currently, almost 94% of surgeons limit administration to a single day (one to four doses). This practice is in accordance with the current thought that longer administration does not decrease the incidence of infection but may actually contribute toward an increase in the number of resistant organisms. However, certain conditions require extended antibiotic administration: delayed operations, oral antibiotics not administered properly, fecal spillage during the procedure, prolonged operations (i.e., >3.5-4 hours), and performance of a rectal resection (e.g., abdominal-perineal operations) [14, 23, 24]. The "ideal" antibiotic prophylaxis would result in few infections and would be inexpensive, easy to administer, and well tolerated by patients. The most popular regimens emulate this ideal by utilizing oral neomycin plus either erythromycin or metronidazole combined with a perioperative parenteral antibiotic (table 5). Although the additional benefit of perioperative antibiotics has not been verified, all recent surveys have shown their popularity. The agents most often added are second-generation cephalosporins (67.7% of cases), drugs that possess activities against both aerobes and anaerobes. Certain other lesser used regimens exhibit variable activities against these organisms, and thus their use should be reconsidered. Those regimens with oral metronidazole or metronidazole plus erythromycin alone do not cover the facultative gram-negative bacteria. However, the addition of an effective parenteral antibiotic or antibiotics tends to cover these organisms and will help protect against postoperative infection. Despite the numerous studies showing the benefit of oral prophylaxis, 54 (11.5%) of our respondents administer only parenteral prophylaxis with drugs that fail to protect against all intestinal microflora. Some of the antibiotic regimens reported in our survey are redundant, using multiple antibiotics with like spectra (table 5). While the length of administration is limited, there remains some concern that this practice may result in the evolution of resistant organisms. It was noteworthy that none of the responding surgeons reported prophylaxis with imipenem/cilistatin. This combination was used in one British trial without oral prophylaxis and was associated with an infection rate of 26.4%, a rate much higher than seen in North American trials [25]. We believe that the use of this combination should remain limited to a therapeutic setting. All surgeons responding to our survey use mechanical bowel cleansing; the most popular preparation is PEG solution (-4 L administered over 3-4 hours the morning of the day before surgery). This regimen has steadily increased in popularity since , while the use of "traditional" procedures has decreased [6, 19]. Although mannitol bowel preparation remains common outside the United States, the rate of its use here has now decreased to <2%. While not previously reported, a large number (28.4%) of our surgeons now routinely use oral sodium phosphate solution with or without bisacodyl in a one- to two-dose regimen before administration of the oral antibiotics. This solution cleanses the bowel by acting as an osmotic purgative and has been shown to be effective for colorectal cleansing without causing any significant clinical problems [26]. It is currently used at our hospital, and its efficacy has been shown to be superior to that of PEG solution in promoting colonic cleansing with relatively small volumes. Although PEG solution has been approved for bowel cleansing before colonoscopy and roentgenographic examinations with barium enemas, oral sodium phosphate solution is also approved for bowel cleansing before surgery [27]. Its use, however, should be limited to those patients without evidence- of kidney disease, congestive heart failure, or other contraindications. The routine use of mechanical cleansing has recently been challenged in several reports from the United Kingdom and Ireland [21, 28-30]. On the basis of observations following emergent and elective colorectal surgery, the researchers concluded that mechanical preparation is not needed to further reduce infection rates provided that effective antibiotics are administered. Irving and Scrimgeour [28] reported an infection rate of 8.3% among 72 patients undergoing elective and emergent

7 CID 1997;24 (April) Preoperative Preparation of the Colon 615 colorectal procedures without bowel cleansing. They believe that oral antibiotics are unnecessary and that bowel cleansing is exhausting to the patient and simply turns solid feces into an uncontrollable liquid. Brownson et al. [29] reported equivalent wound infection rates but higher intraabdominal infection rates among patients mechanically cleansed with PEG solution than among those not mechanically cleansed. Burke and colleagues [30] found that the presence of solid stool in the colon did not appear to increase infection rates. Despite these reports, we believe that further controlled studies are needed before the discontinuance of mechanical preparation can be recommended. Most surgeons responding to the present survey indicated that a grossly clean colon during surgery is an important factor in their choice of bowel preparation. The most commonly used methods of bowel cleansing can be performed on an outpatient basis and can be completed the day before surgery. The increasing use of the sodium phosphate solution shows that an effective mechanical preparation can be obtained relatively quickly without undue stress in the nonobstructed patient. We believe that a clean colon is technically easier to work with; if the colon is clean, the chance of solid fecal spillage decreases, and normal colonic motility returns more quickly in the postoperative course. In 1988, a survey showed that most patients (61%) were admitted to the hospital the day before surgery, and 33% were admitted 2 days before [19]. Very few (3%) received the bowel preparation as outpatients and were admitted on the day of the surgical procedure. The use of outpatient bowel preparation is now increasing, with one report showing the rate of its use increasing from zero to 88% in a 4-year period ending in 1992 [7]. The widespread use of this technique has not resulted in increased rates of infections or other complications [7, 31-33]. Moreover, yearly savings in hospitalization costs of more than $150 million were estimated for the United States alone [7]. Approximately 60% of our survey respondents believe that the home preparation can be as good as the one in the hospital. They indicated, however, that not all patients are candidates for this procedure. Elderly patients or those with contraindicating conditions should not have a home preparation unless they are sufficiently supervised and adequate written and oral instructions are provided. At our institution, home bowel preparation is routinely used and is generally believed to be beneficial provided that the patients are screened and educated about the procedures. The present survey of board-certified colon and rectal surgeons indicates that antibiotic prophylaxis efficacious against both facultative and anaerobic colonic microflora is routinely used and that effective mechanical preparations are used. Although the cost of the complete preparation was of less concern, the surgeons prefer one that is acceptable to the patient, is easy to administer, and results in a low incidence of postoperative infection. These conditions are accomplished by removing gross feces from the intestines by mechanical cleansing, reducing the burden of intraluminal bacteria with administration of oral antibiotics, and providing adequate serum and tissue levels of antibiotics (oral agents with or without parenteral agents) at the time of the operation. No significant changes in antibiotic choices were found from previous surveys. Although the most common mechanical preparation remains oral PEG solution, the use of oral sodium phosphate solution with or without bisacodyl is rapidly increasing. It appears that a combination of a home bowel preparation and both preoperative oral antibiotics (neomycin plus erythromycin or metronidazole) and perioperative parenteral antibiotics (second-generation cephalosporin) is currently the preferred method of prophylaxis for elective colon and rectal surgery. The use of outpatient mechanical preparation is increasing, apparently without an increase in the rate of complications. It is imperative that adequate precautions be taken to maintain the current low level of complications now observed.

8 616 Nichols et al. CID 1997;24 (April) Appendix Bowel Preparation Survey 1995 DEMOGRAPHICS 1) In what year did you become board certified in colon and rectal surgery? 19 2) In which state do you practice? 3) In which size community do you practice? Small (<25,000) Medium (25-100,000) Large (>100,000) 4) In which type of institution do you practice? Check ALL that apply. Community hospital Large medical center Military hospital Teaching hospital Veterans Administration hospital Medical school 5) Approximately how many procedures do you perform each month? Colon Rectum Anus 6) What percentage of your cases are elective or emergent? Elective: % Colon % Rectum % Anus Emergent: % Colon % Rectum,% Anus 7) What percentage of your cases are admitted to the hospital the day of surgery? % Colon % Rectum % Anus BOWEL PREPARATION PROCEDURES Please answer the following questions regarding preoperative bowel preparation for elective, colon and rectal surgical procedures. 1) Which form of mechanical, preparation do you normally use? None "Traditional" using cathartics and enemas over hours / days PEG (polyethylene glycol solution) Mannitol grams Whole-gut lavage liters over hours liters over hours 2) The mechanical preparation is usually started hours before the operative procedure is scheduled. 3) Are there instances where you feel an alternate mechanical preparative method (not your normal method as checked above) should be used? Please list: Completed questionnaire can be sent by FAX to [Page 1 of 3]

9 CID 1997;24 (April) Preoperative Preparation of the Colon 617 Bowel Preparation Survey 1995, 4) Which form ofslat antibiotic prophylaxis do you normally use? Check ALL that apply. None Aminoglycoside Neomycin Kanamycin Clindamycin Erythromycin base Metronidazole Tetracycline 5) The ad antibiotics are started hours before the operative procedure is scheduled. 6) Which form of parenteral, antibiotic, prophylaxis do you normally use? None First generation cephalosporin gram(s) q hrs X doses Cefazolin Second generation cephalosporin gram(s) q hrs X doses Cefotetan Cefoxitin Third generation cephalosporin gram(s) q hrs X doses Ceftizoxime Ceftriaxone gram(s) q hrs X doses 7) The =Mad antibiotics are started hours before the operative procedure is scheduled. 8) Which microorganisms do you feel are most important to protect against in surgical infections following colorectal surgical procedures? Aerobic bacteria (E. coli, Kiebsiella, Enterococcus, etc.) Anaerobes (B. fragilis, Clostridia, etc.) Both are equally important Neither are important Completed questionnaire can be sent by FAX to [Page 2 of 3]

10 618 Nichols et al. CID 1997;24 (April) Bowel Preparation Survey ) At xox institution, which microorganisms are most implicated in surgical infections following colon and rectal procedures? Aerobic bacteria: Anaerobes: Both aerobes and anaerobes are equally implicated. Neither are recovered. 10) In patients which you consider "high risk", are special antibiotic precautions (e.g., longer prophylaxis, additional antibiotics, larger doses) utilized? No Yes: 11) Rate the following factors in influencing your choice of bowel preparation (mechanical & antibiotic): Not Very important Neutral important Cleanliness of operative site Cost Ease of administration Patient acceptance Reduction of bacteria at operative site Reduction of post-surgical infections ) Do you feel that home bowel preparation with the patient admitted to the hospital on the day of surgery is as good as a full hospital based preparation? YES NO (why not) 13) What is the average daily cost per patient day at your institution? $ ADDITIONAL COMMENTS if desired: Thankyou for your valua6k time. We appreciate your input for this important survey! 4 Completed questionnaire can be sent by FAX to [Page 3 of 31

11 CID 1997; 24 (April) Preoperative Preparation of the Colon 619 Acknowledgments The authors thank the American Society of Colon and Rectal Surgeons and the Fellows of the American Society of Colon and Rectal Surgeons. This report would not have been possible without their diligence in filling out the questionnaires. We also thank Whitney T. Michaels, B.S., M.P.H., for her help in reviewing the survey data and analyses. References 1. Finegold SM. Intestinal bacteria. The role they play in normal physiology, pathologic physiology, and infection. Calif Med 1969;110: Nichols RL, Condon RE. Preoperative preparation of the colon. Surg Gynecol Obstet 1971;132: Nichols RL, Condon RE, Gorbach SL, Nyhus LM. Efficacy of preoperative antimicrobial preparation of the bowel. Ann Surg 1972; 176: Nichols RL, Broido P, Condon RE, Gorbach SL, Nyhus LM. Effect of preoperative neomycin-erythromycin intestinal preparation on the incidence of infectious complications following colon surgery. Ann Surg 1973;178: Nichols RL. Bowel preparation. In: Wilmore DW, Cheung LY, Harken AH, Holcroft JW, Meakins JL, eds. Scientific American: Surgery. Vol. 1. New York: Scientific American, 1995: Solla JA, Rothenberger DA. Preoperative bowel preparation: a survey of colon and rectal surgeons. Dis Colon Rectum 1990; 33: Philip RS. Efficacy of preoperative bowel preparation at home. Am Surg 1995;61: Finegold SM. Studies on antibiotics and the normal intestinal flora. Tex Rep Biol Med 1951;9: Nichols RL, Gorbach SL, Condon RE. Alteration of intestinal microflora following preoperative mechanical preparation of the colon. Dis Colon Rectum 1971;14: Nichols RL, Condon RE. Antibiotic preparation of the colon: failure of commonly used regimens. Surg Clin North Am 1971; 51: Bentley DW, Nichols RL, Condon RE, Gorbach SL. The microflora of the human ileum and intra-abdominal colon: results of direct needle aspiration at surgery and evaluation of the technique. J Lab Clin Med 1972; 79: Nichols RL, Condon RE, DiSanto AR. Preoperative bowel preparation: erythromycin base serum and fecal levels following oral administration. Arch Surg 1977;112: DiPiro JT, Patrias JM, Townsend RJ, et al. Oral neomycin sulfate and erythromycin base before colon surgery: a comparison of serum and tissue concentrations. Pharmacotherapy 1985; 5: American Medical Association Division of Drugs and Toxicology. Antimicrobial chemoprophylaxis for surgical patients. In: Drug evaluations annual Chicago: American Medical Association, 1995: Weaver M, Burdon DW, Youngs DJ, Keighley MRB. Oral neomycin and erythromycin compared with single-dose systemic metronidazole and ceftriaxone prophylaxis in elective colorectal surgery. Am J Surg 1986; 151: Kling P-A, Dahlgren S. Oral prophylaxis with neomycin and erythromycin in colorectal surgery: more proof for efficacy than failure. Arch Surg 1989;124: Condon RE, Nichols RL, Bartlett JG. Letter to the editor. Am J Surg 1986; 152: Baum ML, Anish DS, Chalmers TC, Sacks HS, Smith H Jr, 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: Beck DE, Fazio VW. Current preoperative bowel cleansing methods: results of a survey. Dis Colon Rectum 1990; 33: Condon RE, Bartlett JG, Nichols RL, Schulte WJ, Gorbach SL, Ochi S. Preoperative prophylactic cephalothin fails to control septic complications of colorectal operations: results of controlled clinical trial. A Veterans Administration cooperative study. Am J Surg 1979;137: Duthie GS, Foster ME, Price-Thomas JM, Leaper DJ. Bowel preparation or not for elective colorectal surgery. J R Coll Surg Edinb 1990;35: Culver DH, Horan TC, Gaynes RP, et al. Surgical wound infection rates by wound class, operative procedure, and patient risk index. Am J Med 1991; 91(suppl 3B):152S-7S. 23. Gorbach SL. Antimicrobial prophylaxis for appendectomy and colorectal surgery. Rev Infect Dis 1991; 13(suppl 10): Kaiser AB, Herrington JL Jr, Jacobs JK, Mulherin JL Jr, Roach AC, Sawyers JL. Cefoxitin versus erythromycin, neomycin, and cefazolin in colorectal operations: importance of the duration of the surgical procedure. Ann Surg 1983;198: Karran SJ, Sutton G, Gartell P, Karran SE, Finnis D, Blenkinsop J. Imipenem prophylaxis in elective colorectal surgery. Br J Surg 1993;80: Cohen SM, Wexner SD, Binderow SR, et al. Prospective, randomized, endoscopic-blinded trial comparing precolonscopy bowel cleansing methods. Dis Colon Rectum 1994; 37: Physicians' desk reference. 50th ed. Montvale, New Jersey: Medical Economics, Irving AD, Scrimgeour D. Mechanical bowel preparation for colonic resection and anastomosis. Br J Surg 1987; 74: Brownson P, Jenkins SA, Nott D, Ellenbogen S. Mechanical bowel preparation before colorectal surgery: results of a prospective randomized trial [abstract]. Br J Surg 1992;79: Burke P, Mealy K, Gillen P, Joyce W, Traynor 0, Hyland J. Requirement for bowel preparation in colorectal surgery. Br J Surg 1994;81: Frazee RC, Roberts J, Symmonds R, Snyder S, Hendricks J, Smith R. Prospective, randomized trial of inpatient vs. outpatient bowel preparation for elective colorectal surgery. Dis Colon Rectum 1992; 35: Handelsman JC, Zeiler S, Coleman J, Dooley W, Walrath JM. Experience with ambulatory preoperative bowel preparation at the Johns Hopkins Hospital. Arch Surg 1993;128: Lee EC, Roberts PL, Taranto R, Schoetz DJ Jr, Murray JJ, Coller JA. Inpatient vs. outpatient bowel preparation for elective colorectal surgery. Dis Colon Rectum 1996; 39: