Are Prophylactic Antibiotic Agents Indicated in Anorectal Surgery for Patients with Human Immunodeficiency Virus?

Abstract

Background:

With increased survival among patients with human immunodeficiency virus (HIV), surgeons have been seeing more cases of anal dysplasia and cancer. There is, however, no data on the incidence of surgical site infections (SSIs) in HIV-positive patients undergoing elective anorectal procedures, nor on the administration of prophylactic antibiotic agents. We reviewed a HIV-positive population that has undergone elective anorectal biopsy of areas of dysplasia observed on office anoscopy to assess the need for antibiotic prophylaxis.

Patients and Methods:

A retrospective chart review was performed of all HIV-positive patients seen as outpatients in the Colorectal Surgery Division from 2007–2016. Demographics, dates of surgery and follow-up, antibiotic prophylaxis, and pre-operative CD4 count and HIV viral load were recorded for 229 patients. Post-operative examination notes were reviewed to determine the presence of SSIs. The proportion of patients who received prophylaxis was assessed and the SSI rate was calculated.

Results:

Surgical site infections occurred in 2 of 237 (0.8%) cases without antibiotic prophylaxis and in none of the 38 cases with prophylaxis. This infection rate was found to be lower than that of the general surgery population, with no statistical difference from hemorrhoidectomy patients without HIV. One SSI occurred in a 51-year-old male with a pre-operative CD4 count of 612 per microliter and viral load of zero. Another occurred in a 57-year-old female with an unknown CD4 count and viral load. A χ² analysis showed the incidence of SSIs in the groups with and without prophylaxis was not significantly different (p = 0.563).

Conclusion:

Surgical site infection rates in HIV-positive patients undergoing biopsies for anal dysplasia were similar to patients without HIV undergoing similar minor anorectal procedures, and no difference was noted in the rate of SSI with the administration of prophylactic antibiotic agents. We do not recommend routine use of prophylactic antibiotic agents in this population.

Human immunodeficiency virus (HIV) has been a challenge to healthcare providers since its discovery. With improving awareness, health outreach, and medical therapy, doctors treat more HIV-positive patients than ever before. The most recent data from the U.S. Centers for Disease Control and Prevention (CDC) in 2014 state that approximately 44,073 people are diagnosed with HIV in the United States annually, whereas there are approximately 1.2 million people currently living with HIV in the United States [1]. As highly active antiretroviral therapy (HAART) improves survival among people with an HIV infection, surgeons have been seeing more HIV-positive patients with benign and malignant anorectal pathologies, including anal dysplasia and cancer. Furthermore, anal dysplasia has become increasingly prevalent in this population, and is often treated by elective excision and fulguration to prevent it from progressing to squamous cell carcinoma of the anus (SCCA) [2].

The risk of surgical site infections (SSIs) is an important complication to consider for HIV-positive patients. Surgical site infections increase morbidity, mortality, the length of hospital stay, and thus the cost of surgery [3,4]. Most SSIs from elective anorectal procedures are classified as clean-contaminated because the operative procedure enters into a cavity of the body under elective and controlled circumstances. For the general population, the SSI rate for this class is 3%–11% [5]. In patient samples specific for HIV-positive individuals, one study has reported an SSI rate of 9.5% across all surgical HIV-positive patients, almost twice as high as non-HIV–positive patients [6]. However, the study did not examine CD4 counts or viral loads for their patients. Another study suggests an increased risk of SSIs in the general surgical population only when pre-operative CD4 counts are less than or equal to 200 cells per microliter [7]. Yet, there are no studies on SSIs focusing specifically on elective anorectal procedures in the patient with HIV.

Furthermore, studies have presented conflicting results concerning the CD4 count at which SSI risk is increased in HIV-positive patients. Whereas older studies found no predictive value in perioperative T cell counts on surgical outcome [8,9], more recent studies have found HIV-positive patients undergoing surgical procedures with CD4 counts of less than 50 per microliter or with HIV viral loads greater than 10,000 c/mL to have an increased risk for surgical site complications [10,11]. Viral loads greater than 30,000 c/mL present an even greater risk for surgical site complications. However, there are no data to suggest that HIV-positive patients with higher CD4 counts and lower viral loads are at a different risk of developing SSIs compared to non-HIV–positive patients, nor are there data to suggest prophylactic antibiotic agents should be administered to this population. Current studies on multiple surgical procedures (cholecystectomy, hip arthroplasty, spine surgery, herniorrhaphy, and coronary artery bypass grafting) have shown that CD4 cell count is not an independent risk factor for post-operative surgical complications in HIV-positive patients [12]. Human immunodeficiency virus infection status did not represent an independent risk factor for post-operative complications; American Society of Anesthesiologists (ASA) risk class was the most crucial risk factor [13].

Antibiotic agents are costly and can have serious side effects, including drug reactions, opportunistic infections, and antibiotic resistance [14 –16]. The identification of the risk of SSIs is crucial for the effective targeting of a population for prophylactic antibiotic agents. As noted previously, the SSI risk for HIV-positive patients had been explored for multiple surgical procedures, including appendectomy, arthrotomy or arthroscopy, bowel resection, cholecystectomy, cardiothoracic procedures, hernia repair, hysterectomy, hip or knee replacement, laparoscopy or laparotomy, and mammoplasty, but no study has been published in regards to anorectal surgery.

Typically, prophylactic antibiotic agents are not indicated for outpatient anorectal surgery, and in patients without HIV undergoing outpatient excisional hemorrhoidectomy, they did not affect the rate of SSIs [17,18]. Prior studies show a significant decrease in incision healing from anorectal surgery in HIV-positive patients with CD4 counts lower than 50. Yet, the study population does not distinguish between elective and non-elective cases, and many of the surgeries were done on actively infected patients [19].

The role of prophylactic antibiotic agents in elective outpatient anorectal surgery for anal dysplasia in HIV-positive patients is unclear. Currently, the American Society for Colon and Rectal Surgery (ASCRS) does not address specifically pre-operative antibiotic agents for anorectal procedures [20]. We hypothesized that the use of prophylactic antibiotic agents is not necessary in the surgical management of HIV-positive patients undergoing outpatient anorectal surgery for patients with a CD4 count above 50 per microliter, regardless of their viral load. The purpose of this study is to review our HIV-positive patient population that has undergone outpatient anorectal surgery for dysplasia and determine if prophylactic antibiotic agents are necessary. CD4 cell counts, viral loads, and administration of prophylactic antibiotic agents were recorded for each patient, and data on surgical wound infection were assessed.

Patients and Methods

Study design

After Institutional Review Board approval was obtained, we performed a retrospective chart review of all HIV-positive patients undergoing outpatient anorectal surgery with post-operative follow-up seen in the colorectal surgery clinic from 2007 to 2016. Surgeries were performed by the same two board-certified colorectal surgeons throughout the entire study period.

Basic demographics and clinical data such as date(s) of surgery, follow-up visits, pre-operative CD4 counts, and viral loads within six months prior to surgery, and peri-operative antibiotic prophylaxis were recorded. Two independent reviewers examined the post-operative outpatient examination notes within 30 days of surgery to determine if an SSI was present. The reviewers searched for key words such as, “redness, induration, erythema, tenderness, purulent drainage, and pain” to determine if an infection was present. They also looked to see if the patient reported a fever, or if antibiotic agents were prescribed during that visit, indicating an infection. This corresponded to the CDC definition for classification of an SSI, which is an infection within 30 days of the operation that is marked by inflammation, redness, and purulence, and/or fever.

Patients were stratified according to their CD4 counts. In each strata, the proportion of patients who received prophylactic antibiotic agents was assessed. Similar analysis was completed for patients for whom CD4 count was unavailable. Cases for which prophylactic antibiotic agents were administered and those for which they were not were also analyzed separately to study the type of antibiotic and SSI rate, independent of CD4 count. Microsoft Excel (Microsoft Corporation, Redmond, WA) was used to perform χ² analysis, with a p value of 0.05 used to determine significance. A post hoc power analysis was performed against accepted SSI rates in the literature for clean-contaminated cases of non-HIV–infected patients.

Patient population

All patients were referred to our center from the primary physician treating their HIV infection. Patients were included in the study if they underwent surgical biopsy for anal lesions, and had follow-up in our clinic of at least one month. Exclusion criteria were patients with un-documented prophylactic antibiotic status, patients taking steroids, patients being treated with immunosuppression, or patients with active infectious or neoplastic disease. Patients did not complete any bowel preparation prior to surgery. In surgery, the patient was placed in the prone jackknife position, and the anal verge was cleaned with Betadine^® solution (Purdue Pharma LP, Stamford, CT). Antibiotic prophylaxis was administered intravenously prior to incision. The anal canal is not prepared with Betadine prior to use of the anoscope. After surgery, the patient is instructed to take stool softeners or fiber to keep their stool consistency soft, and soak in a Sitz bath for any discomfort. They are given a prescription for pain medications unless otherwise contraindicated.

Results

Initially, 229 HIV-positive patients were selected who underwent 361 procedures from the colorectal surgery clinic at Hahnemann University Hospital from January 1, 2007 through June 24, 2016. After exclusion criteria were applied, 161 patients were included who underwent 275 outpatient anorectal procedures. The 161 patients analyzed were aged 17 to 64 years with a mean age of 41.6 years and a median age of 43 years. Of the 275 procedures, 273 had no SSIs and 2 had SSIs. Patient age and gender for the 275 procedures are presented in Table 1. Among the 275 procedures, the CD4 count range is presented in Table 2 and the viral load range is presented in Table 3.

Table 1.

Age and Gender of Patients Undergoing Surgery

Gender	Number of procedures (%)
Male	242 (88)
Female	33 (12)
Age at procedure, y
17–29	47 (17.1)
30–39	69 (25.1)
40–49	85 (30.9)
50–59	65 (23.6)
>60	9 (3.3)

Table 2.

CD4 Count of Patients with and without Pre-Operative Antibiotic Prophylaxis

	<50 per microliter	50–200 per microliter	>200 per microliter	Unknown
No. of cases (% total)	5 (1.8%)	15 (5.5%)	98 (36%)	157 (57%)
No. with antibiotic agent (% within strata)	3 (60%)	2 (13%)	14 (14%)	19 (12%)

Table 3.

Viral Load Stratification of Patients with Antibiotic Prophylaxis

	< 500 c/mL	500–30,000 c/mL	>30,000 c/mL	Unknown
No. of cases (% total)	87 (32%)	9 (3.3%)	13 (4.7%)	166 (60%)
No. with antibiotic agent (% within strata)	12 (14%)	1 (33%)	4 (31%)	21 (13%)

In total, 38 (13.8%) procedures were performed with prophylactic antibiotic agents. Most often, patients received cefazolin or ertapenem (Table 4). None of the patients who received pre-operative antibiotic agents had any evidence of an SSI. There was one case of Clostridium difficile in a patient receiving antibiotic prophylaxis (0.4% of total cases, 2.6% of cases receiving antibiotics). Of the 237 cases without prophylaxis, 235 cases (99.2%) did not result in a SSI. The two cases without prophylaxis and with documented SSIs yielded an incidence of 0.8%. The two cases of SSIs were in two separate patients. A χ² analysis between patients who did and did not receive prophylactic antibiotic agents showed no difference in rate of SSI (p = 0.563).

Table 4.

Percentage of Antibiotic Received for Pre-Operative Prophylaxis

Cefazolin	42.10%
Ertapenem	36.80%
Metronidazole	13.20%
Cefazolin and metronidazole	5.30%
Clindamycin and metronidazole	2.60%

Case 1 occurred in a 47-year-old male with a past medical history of HIV, obesity, asthma, and obstructive sleep apnea. The patient had a total of four procedures from 2007 to 2012: three excisions and fulgurations of anal condylomas and one hemorrhoid artery ligation. An SSI resulted after the fourth surgery, for which he did not receive antibiotic prophylaxis. His CD4 count prior to surgery was 612 and he had no detectable viral load. The biopsy results of the excision were negative for anal intra-epithelial neoplasia, so no human papillomavirus (HPV) status was reported. The infection was marked by wound inflammation and yellow discharge from the anal area at his follow-up appointment 19 days post-operation.

Case 2 occurred in a 54-year-old female with past medical history of HIV, hypertension, and gastroesophageal reflux disease (GERD). The patient had a total of three excision and fulguration procedures from 2012 to 2014. The second surgery resulted in an SSI, marked by purulent discharge and severe pain with a lump on the right side of the anus. Cephalexin was prescribed. Follow-up was completed 12 days post-operation. Prophylactic antibiotic agents were not given for this procedure. The biopsy result from the third procedure showed high grade anal intra-epithelial neoplasia (HGAIN) but no serotype was performed on this sample. The prior excision and fulguration procedure was performed one year prior and did not result in an SSI. That surgical sample showed HGAIN, and was positive for low risk (6 or 11) and high risk (16, 18, 31, 33, 35, 39, 45, 51, 56, 58, 59 or 68) HPV strains.

Discussion

The advent of HAART has decreased the mortality associated with HIV, and surgeons are seeing more patients for anorectal procedures. The decision of whether to administer antibiotic prophylaxis is a risk versus reward situation that must be evaluated by appropriate data. There are no published rates of SSIs specific to HIV-positive patients undergoing excision and fulguration with respect to the administration of antibiotic prophylaxis. In a meta-analysis of open versus closed hemorrhoidectomy, combining 11 studies yielded an SSI rate of 0.5% for open hemorrhoidectomy versus 1.7% for closed hemorrhoidectomy, with 663 patients in each patient group. Human immunodeficiency virus status or peri-operative antibiotic agents were not noted [21]. A separate larger retrospective review aimed at antibiotic prophylaxis in excisional hemorrhoidectomy found a similar rate of SSI at 1.4% [18]. In a study of SSI healing in HIV-positive patients after excision and fulguration of condyloma acuminata, they note the percentage of wounds healed after three months was 90% [19]. In a prospective multi-center Italian study of HIV-positive patients across all areas of surgery, their overall SSI rate was reported at 9.5%, with a reported rate of 8.5% (confidence interval [CI] 1.1–15.9) in gastrointestinal surgery [6]. In another retrospective study of HIV-positive patients, they report a 93% incisional SSI rate for HIV-positive patients undergoing peri-anal surgery, but only a 9% SSI rate of HIV-positive patients undergoing skin tumor removal [7]. Many of these studies did not comment on the administration of peri-operative antibiotic agents or time to follow-up. Finally, in a recent retrospective review of the National Surgical Quality Improvement Program (NSQIP) database, a SSI rate of 3.94% for clean-contaminated, and 4.75% for contaminated cases was noted [5].

Of the 275 procedures evaluated, we found the incidence of SSI to be 0.8% in HIV-positive patients undergoing excision and fulguration for condyloma acuminata. The study is not powered adequately to determine if this rate is different from that observed in hemorrhoidectomy in non-HIV–infected patients (1.4% SSI, 7.1% powered). However, the study is powered enough to detect a difference in SSI from all other samples mentioned above, with the lowest comparator being clean-contaminated cases across all surgical HIV-positive patients status unknown (3.95% SSI, 94.1% powered). Given this, our data confirm that the SSI rate in HIV-positive patients undergoing excision and fulguration is less than that of patients undergoing clean-contaminated cases, and it may be the same as non-HIV–infected patients undergoing hemorrhoidectomy. This suggests that HIV-positive patients undergoing excision and fulguration are not at an increased risk for SSIs compared to other patients undergoing hemorrhoidectomy and those patients do not receive antibiotic prophylaxis routinely.

The decision for prophylactic antibiotic agents was ultimately left to the operating surgeon. There was no significant difference in prophylaxis rate between those with CD4 counts below 500, and those with unknown CD4 counts. Furthermore, the median date of service was nearly identical between two groups (November 10, 2012 for those without prophylaxis, September 20, 2012 for those with prophylaxis), suggesting more recent publications did not have a significant impact on decisions for prophylaxis [7,18].

CD4 count is an important marker for HIV disease progression. Our study found that three of five patients with CD4 counts below 50 per microliter were administered pre-operative antibiotic agents. Studies generally agree that patients with counts below 50 per microliter are at a higher risk for developing SSIs and antibiotic and antifungal agents should be started pre-operatively as prophylaxis [7]. Interestingly, although none of the patients who received prophylactic antibiotic agents developed an SSI, one SSI occurred in a patient with a CD4 count greater than 200 per microliter, and one in a patient with unknown CD4 count. No SSIs were found among patients with CD4 counts below 50 per microliter, although there were only five patients in this range and 60% of them received prophylactic antibiotic agents. Given this distribution, CD4 counts may not be the only or the most influential factor in SSI risk assessment among HIV-positive patients [12]. In a prospective study across all surgical HIV-positive patients, only hepatitis C virus (HCV) status was noted for an increase in risk for post-operative SSIs. In this study, neither CD4 count nor pre-operative antibiotic prophylaxis had a significant effect on post-operative SSI rates [6]. It was not possible to collect HCV data on most of the patients in our study because the majority of their care was provided outside the purview our electronic medical records.

Viral load has also been shown to be an important marker for disease progression in HIV-positive patients. Most procedures (31.6%) with known viral loads were completed with patient viral loads less than 500 c/mL; SSI risk is lower in this bracket, representing an optimal time for surgery [6]. Only 4.72% of procedures were completed with patient viral loads above 30,000 c/mL. Whereas 13.8% of procedures with patient viral loads below 500 c/mL were administered prophylactic antibiotics, 30.8% of procedures with viral loads above 30,000 c/mL were given pre-operative antibiotic agents. The one SSI in our population for which viral load was available, however, had no detectable viral load.

Current studies suggest routine administration of prophylactic antibiotic agents for certain anorectal cases does not improve outcomes [17], and it is not our practice to administer them routinely in these cases. Given this, our data suggest prophylactic antibiotic agents may not be required for patients with CD4 counts above 50. Whereas no SSIs were noted in patients with CD4 counts below 50, prior studies suggest this group to be at risk of SSI healing complications. In a study specific for SSI healing in HIV-positive patients across all anorectal surgery, patients undergoing biopsies for condylomas were found to have the highest percentage of healed SSIs at three months at a rate of 90%. Yet, the only prognostic factor found for SSI healing was CD4 count, with patients having CD4 counts above 50 having an odds ratio of 5.07 compared to those below 50 (p = 0.010) [19].

The primary limitation of this study was the number of patients enrolled. Whereas the study was powered adequately to detect a difference in overall SSI rate between HIV-positive patients and the general population, it was not powered adequately to detect a difference between groups of CD4 count or viral load. In order to detect a doubling of SSI rate between patients with a CD4 count between 50–200 per microliter and those with less than 50 per microliter, a sample size of 500 patients in each group is necessary to obtain 86.1% power. Our data suggest that CD4 count does not influence SSI rate from elective anorectal excision and fulguration, which mirrors literature data on SSI healing rates for this procedure in HIV-positive patients [19].

Regular antibiotic agents are costly and have side effects, which should encourage accurate targeting of use. A cost analysis of administering prophylactic antibiotic agents to all patients could show more definitively the costs of such a protocol [4]. It is worth noting that the two SSIs observed in our patient population were easy to treat and caused minimal morbidity and no mortality, further supporting the conclusion that pre-operative antibiotic agents may not be necessary for these patients. A future aim of this study is to review the procedures with patient CD4 counts above 50 per micoliter and viral loads less than 500 c/mL who received prophylactic antibiotic agents and assess other patient risk factors for SSIs.

Conclusion

Based on our study, SSI rates in HIV-positive patients undergoing anorectal procedures were substantially lower than those of the general surgical populations having laparotomies and other major clean-contaminated cases. Furthermore, the SSI rate of HIV-positive patients is similar to that of non-HIV–infected patients undergoing similar minor anorectal procedures, with no difference noted when prophylactic antibiotic agents were administered. Complications from administration of pre-operative antibiotics, such as the rate of Clostridium difficile infection, was higher than the SSI rate in patients who did not receive pre-operative antibiotic agents. Antibiotic agents are costly and have many side effects, and we do not recommend routine use of prophylactic antibiotic agents for outpatient anorectal procedures in HIV-positive patients with CD4 counts above 50 per microliter. More data are needed to determine if viral load and CD4 count affect SSI rates in anorectal surgery.

Footnotes

Acknowledgment

This study was presented by S.A.P. as a podium presentation at the 12th Annual Academic Surgical Congress of the Association of Academic Surgery on February 8, 2017 in Las Vegas, Nevada.

Author Disclosure Statement

No competing financial interests exist.

References

McNair

, Whistance

, Forsythe

, et al. Core outcomes for colorectal cancer surgery: A consensus study. PLoS medicine. 2016; 13:e1002071.

Long

, Menon

, Bastawrous

, Billingham

. Screening, surveillance, and treatment of anal intraepithelial neoplasia. Clin Colon Rect Surg, 2016; 29:57–64.

Pollard

, Newman

, Barlow

, et al. Deep wound infection after proximal femoral fracture: Consequences and costs. J Hosp Infect, 2006; 63:133–139.

Kirkland

, Briggs

, Trivette

, et al. The impact of surgical-site infections in the 1990s: Attributable mortality, excess length of hospitalization, and extra costs. Infect Control Hosp Epidemiol, 1999; 20:725–730.

Ortega

, Rhee

, Papandria

, et al. An evaluation of surgical site infections by wound classification system using the ACS-NSQIP. J Surg Res, 2012; 174:33–38.

Drapeau

, Pan

, Bellacosa

, et al. Surgical site infections in HIV-infected patients: Results from an Italian prospective multicenter observational study. Infection, 2009; 37:455–460.

Zhang

, Liu

, Zhang

, et al. Prevention and treatment of surgical site infection in HIV-infected patients. BMC Infect Dis, 2012; 12:115.

Safavi

, Gottesman

, Dailey

. Anorectal surgery in the HIV+ patient: Update. Dis Colon Rectum, 1991; 34:299–304.

Beck

, Jaso

, Zajac

. Surgical management of anal condylomata in the HIV-positive patient. Dis Colon Rectum, 1990; 33:180–183.

10.

Horberg

, Hurley

, Klein

, et al. Surgical outcomes in human immunodeficiency virus-infected patients in the era of highly active antiretroviral therapy. Arch Surg, 2006; 141:1238–1245.

11.

Morandi

, Merlini

, Salvaggio

, et al. Prospective study of healing time after hemorrhoidectomy: Influence of HIV infection, acquired immunodeficiency syndrome, and anal wound infection. Dis Colon Rectum, 1999; 42:1140–1144.

12.

King

Jr ., Perkal

, Rosenthal

, et al. Thirty-day postoperative mortality among individuals with HIV infection receiving antiretroviral therapy and procedure-matched, uninfected comparators. JAMA Surg, 2015; 150:343–351.

13.

Jones

, Schechter

, Smith

, Rose

. Is HIV infection a risk factor for complications of surgery?. Mt Sinai J Med, 2002; 69:329–333.

14.

Michaelidis

, Fine

, Lin

, et al. The hidden societal cost of antibiotic resistance per antibiotic prescribed in the United States: An exploratory analysis. BMC Infect Dis, 2016; 16:655.

15.

Ventola

. The antibiotic resistance crisis: Part 1: Causes and threats. P T, 2015; 40:277–283.

16.

Cosgrove

. The relationship between antimicrobial resistance and patient outcomes: Mortality, length of hospital stay, and health care costs. Clin Infect Dis, 2006; 42(Suppl 2):S82–89.

17.

Shaw

, Ternent

. Perioperative management of the ambulatory anorectal surgery patient. Clin Colon Rectal Surg, 2016; 29:7–13.

18.

Nelson

, Champagne

, Rivadeneira

, et al. Prophylactic antibiotics for hemorrhoidectomy: Are they really needed?. Dis Colon Rectum, 2014; 57:365–369.

19.

Lord

. Anorectal surgery in patients infected with human immunodeficiency virus: Factors associated with delayed wound healing. Ann Surg, 1997; 226:92–99.

20.

Ternent

, Fleming

, Welton

, et al. Clinical practice guideline for ambulatory anorectal surgery. Dis Colon Rectum, 2015; 58:915–922.

21.

Bhatti

, Sajid

, Baig

. Milligan-Morgan (open) versus Ferguson Haemorrhoidectomy (closed): A systematic review and meta-analysis of published randomized, controlled trials. World J Surg, 2016; 401509–1519.