Anorectal Abscess

Abstract

Anorectal abscesses are a common colorectal emergency. The hallmark of treatment is obtaining source control while avoiding injury to the underlying sphincter complex. Understanding the anatomy of an anorectal abscess is critical to planning the appropriate drainage strategy and decreasing the risk of complex fistula formation. Use of antibiotics should be reserved for those with extensive cellulitis, signs of systemic infection, or patients who are immunocompromised. Whether antibiotics prevent future fistula formation is an area of active research. Primary fistulotomy at time of the index drainage is controversial; however, there may be situations where it is appropriate. It is important to counsel patients that after effective drainage of an anorectal abscess, they have a 30 to 50% chance of developing an anal fistula that will then require further treatment.

Anatomy and Pathophysiology

Most anorectal abscesses, and the focus of this review, are cryptoglandular in nature. The cryptoglandular theory proposes that anorectal abscesses result from blocked intramuscular anal glands. ¹ These glands, which are located circumferentially in the anal canal at the level of the dentate line, typically drain into the crypts of Morgani. ^{1 2} When the gland becomes obstructed, stasis leads to bacterial overgrowth, and an abscess forms if the gland is unable to decompress into the anal canal.

As pressure builds in the blocked gland, bacteria are forced down the path of least resistance into the potential space between the internal and external sphincter. Thus, a cryptoglandular anorectal abscess initially begins as an intersphincteric abscess ( Fig. 1 ). ² From there, if the infection extends downward and infiltrates the subcutaneous tissue surrounding the anus, a perianal abscess forms. This is the most common manifestation of an anorectal abscess. ³ The second most common presentation is an ischiorectal abscess, which occurs when the infection instead penetrates the external anal sphincter and extends laterally into the ischiorectal fossa. ⁴

Fig. 1.

Anatomy of the anal canal and location of the anorectal abscesses.

Rare upward extension from the intersphincteric space can create a supralevator abscess. In addition to these typical forms, there are also abscesses of the deep postanal space, horseshoe abscesses, and atypical supralevator abscesses that will be discussed later. It is critically important to understand the underlying anatomy of anorectal abscess as this determines the surgical approach. ⁵

Other causes of anorectal abscesses, which are beyond the scope of this review but should be considered in the appropriate clinical context, include inflammatory bowel disease, trauma (including recent anorectal procedures), neoplasms, and rare infections such as tuberculosis.

Epidemiology and Risk Factors

Defining the true incidence of anorectal abscess is challenging as many abscesses spontaneously drain prior to a patient seeking care or are treated in the outpatient setting. Extrapolating from several population studies of anal fistulas, anorectal abscesses are estimated to affect 68,000 to 96,000 people annually in the United States. ⁶ A majority of these patients are in their fourth decade of life and there is a high male preponderance with a male-to-female ratio of 2 to 3:1. ^{3 7 8 9}

Risk factors for the development of a cryptoglandular anorectal abscess include recent smoking, obesity, and diabetes mellitus, especially in patients with poor glycemic control. ^{7 10 11 12} Because of the high co-occurrence of anorectal abscess and diabetes, the World Society of Emergency Surgery suggests checking a serum glucose and hemoglobin A1c in these patients to identify undiagnosed diabetes melitus. ¹³

Clinical Manifestations

The majority of patients present with perirectal pain, although the manifestations of an anorectal abscess will depend in part on its anatomical location. ¹⁴ The pain is often independent of defecation, but patients can report exacerbation by bowel movements. Other common symptoms include swelling and erythema. Rarely, acute urinary retention can result. This is caused by either a direct mechanical obstruction or the surrounding inflammation from the abscess. ¹⁵ The patient may exhibit systemic disturbances such as fever, tachycardia, and leukocytosis. If present, these signs should trigger a heightened level of attention from the clinician, as it can portend a more serious or extensive infection.

The anatomy of the abscess will dictate the clinical presentation. Because perianal abscesses are limited to the plane that surrounds the anal verge, they are generally small and easily visible, and usually do not result in significant systemic symptoms. In contrast, ischiorectal abscesses develop within the larger ischiorectal fossa, allowing them to become quite large with less outward signs of infection. Consequently, patients are more likely to have systemic signs of infection. The external examination usually reveals an indurated, tender, and erythematous area on the medial buttock. Pure intersphincteric abscesses will typically present with perirectal pain, but without any external induration or fluctuance. ¹⁶ Often patients are unable to tolerate a digital rectal examination and require an examination under anesthesia. Similarly, supralevator abscesses may have an unrevealing external examination. ¹⁷ A careful digital rectal examination may demonstrate a tender fluctuance at or above the level of the anorectal ring, but this can be a subtle finding. Since many patients will endorse pelvic or even abdominal pain, these abscesses will frequently be diagnosed via radiographic imaging.

Differential Diagnosis

There are several nonanorectal infectious entities that can present similarly to an anorectal abscess. The distinction is that these processes have no relation to the anal canal. They include simple subcutaneous abscesses, carbuncles, and infected sebaceous cysts. A pilonidal abscess can sometimes be misdiagnosed as an anorectal abscess but is distinguished based on its midline location in the gluteal cleft. Finally, another mimic is hidradenitis suppurativa, which can be differentiated by its combination of deep-seated nodules, draining tracts, and fibrotic scars.

Other processes that result in perianal pain must also be considered. These include a thrombosed external hemorrhoid, acute anal fissure, perianal Crohn's flare, and malignancy. A thrombosed external hemorrhoid should be easily identifiable on external examination as a purple or black swelling associated with enlarged external and internal hemorrhoidal tissue. An acute anal fissure results in a sharp pain with defecation with a period of throbbing or burning that can last several minutes to hours following a bowel movement. With gentle effacement of the buttocks during the physical examination, a tear in the anoderm with sentinel skin tag can be visualized and any tenderness is limited to this area. Finally, it is important to note abnormal or waxy-appearing perianal skin, inflammatory skin tags, or multiple involved areas that can suggest a diagnosis of perianal Crohn's disease. One population-based study found that 2.9% of all patients presenting with a new anorectal abscess were subsequently diagnosed with Crohn's disease. ¹⁸

Diagnostic Imaging

The use of diagnostic imaging for most anorectal abscesses is not necessary. ¹⁹ In fact, the lack of finding on a computed tomography (CT) scan does not rule out an abscess. In one retrospective review, a CT scan obtained less than 48 hours prior to drainage was negative in 23% of patients with a confirmed anorectal abscess. ²⁰ However, CT scans can be helpful in patients presenting with perirectal pain and concern for an abscess without external examination findings. A CT scan may also be considered in patients with a suspected abscess but without intraoperative findings of abscess in the operating room to rule out supralevator infection.

More advanced imaging modalities such as magnetic resonance imaging (MRI) can assist in investigating complex abscesses and fistulas; however, it is not useful in the emergent setting due to time and availability. ²¹ Specifically, in cases of complex supralevator abscesses, an MRI can clarify the precise anatomical location, which can inform the appropriate drainage route. ²² Other modalities, such as transperineal ultrasound, have a sensitivity and specificity of 100 and 94%, respectively, for the detection of perianal abscesses. ²³ However, patient discomfort and operator dependency limit widespread adoption. ²⁴

Treatment

The primary treatment of an anorectal abscess is incision and drainage. Before proceeding, it is recommended that information about baseline anal sphincter function, history of anorectal operations, obstetric history, use of blood thinners, and associated urinary or gynecological pathology be obtained. ¹⁹

Location and Positioning

Most perianal abscesses can be drained at the bedside using just local anesthetic. ²⁵ Small superficial ischiorectal abscesses can also be managed this way; however, more extensive abscesses in this location or in patients on blood thinners or with signs of sepsis should be drained in the operating room to ensure that adequate source control with hemostasis is obtained. One retrospective analysis of 458 patients found that abscess drainage in the operating room led to a decreased rate of abscess recurrence. ²⁶

For perianal and small ischiorectal abscesses that are being treated in the outpatient setting, the patient is usually placed in either left or right lateral decubitus or, if an appropriate examination table is available, in the knee–chest position. For larger ischiorectal abscess or those that need to be approached transanally (e.g., intersphincteric), the patient is brought to the operating room and placed into either a prone jackknife or high lithotomy position. Wherever the procedure is performed, an adequate source of lighting is important and many providers find headlamps to be useful.

Technique for Perianal and Ischiorectal Abscess

How an anorectal abscess is drained is determined by its relationship to the sphincter complex. The drainage of a perianal or ischiorectal abscess should be performed as close to the anus as possible rather than over the area of maximum fluctuance as is the standard for other cutaneous abscesses. Placing the drainage incision close to the anus minimizes the length of a potential fistula should one develop in the future. A hemostat or blunt probe can be used to gently explore the cavity and ensure complete drainage by breaking up all loculations. However, mechanical distribution of all fibrous tissue within the abscess cavity should be avoided, as this can injure the pudendal nerve branches that spread outward from the anal verge. ²⁷

If no apparent fluctuance is visible externally, an 18-guage needle can be used to aspirate the region where an abscess is suspected based on either imaging or rectal examination. A caveat to this technique is that the abscess fluid may be too thick for the needle, misleading the clinician. To prevent premature closure of the drainage site, a segment of skin should be excised, in either an ellipse or a cruciate with the corners removed. The cavity can be irrigated with saline. Hemostasis can be obtained with direct pressure, silver nitrate, cautery if available, or using lidocaine with epinephrine.

Technique for Intersphincteric Abscess

Once properly sedated and positioned, a digital rectal examination can reveal an area of fluctuance that protrudes into the anal canal. The mucosa overlying the fluctuant area is opened, which exposes the fibers of the underlying internal sphincter muscle. A Kelly forceps is passed through the internal sphincter into the abscess cavity and should result in a pronounced efflux of pus. The instrument is gently spread to create an aperture for drainage with care not to damage the surrounding sphincter complex. Some practitioners argue to lay open the abscess and divide the internal sphincter up to the level of the dentate line or higher if necessary. ²⁸

Technique for Supralevator Abscess

The underlying anatomy and etiology of a supralevator abscess is essential to understand since it will dictate the correct surgical approach. When these abscesses originate from a superior extension of an intersphincteric abscess, they should be treated with drainage via the rectum to avoid inadvertently creating a complex suprasphincteric fistula. In other situations, the suprasphincteric abscess arises from an extension of an ischiorectal abscess and should be drained externally through the perianal skin. Zinicola et al describe this as the “skeletal muscle rule”; if the infectious process has not passed through the skeletal muscle (external anal sphincter and levator ani), it should be drained inward, whereas if it does, the abscess should be drained outward. ²⁹ In this clinical scenario, MRI can be helpful to determine the appropriate surgical management. ^{22 30} A third scenario is a supralevator abscess that is a result of an intra-abdominal infectious process such as perforated diverticulitis. In this case, transabdominal percutaneous drainage should be pursued if technically feasible. However, depending on the abdominal process and stability of the patient, source control may require transabdominal surgery.

Technique for Horseshoe Abscess

This distinct subset of anorectal abscess begins as an infection of the deep postanal space and then progresses with a unilateral or bilateral extension into the ischiorectal fossae. ³¹ The deep postanal space is located between the tip of the coccyx and the anal orifice, deep to the superficial external anal sphincter and superficial to the levator muscles. ³² The key to the management of a horseshoe abscess is adequate drainage of the deep postanal space. This is accomplished in the operating room with the modified Hanley procedure, which involves division of the anococcygeal ligament and creation of counter incisions overlying the ischiorectal fossae. Several setons are then used to encourage drainage. These setons are typically placed from both counter incisions to the midline. Then the fistula track from the deep postanal space to the posterior midline of the anal canal is identified and encircled with a seton ( Fig. 2 ). ³³ The setons in the counter incisions are sequentially removed in the outpatient setting. Once the inflammatory process has resolved and the fistula tract has matured, a definitive fistula procedure is performed. ³⁴ For deep postanal space infections without horseshoe extension, an intersphincteric approach has been described with a high success rate. This involves making an intersphincteric internal incision and extending this cephalad to enter into the deep postanal space to drain the abscess. The internal opening on the surface of the internal sphincter is then closed. ³⁵

Fig. 2.

Following drainage of the deep post anal space infection via the modified Hanley procedure.

Use of Packing

Several observational studies have shown that conventional packing increases both pain and treatment costs for patients. ^{36 37} In a randomized study of 50 patients, there was no difference in healing rates, abscess recurrence, or fistula formation between patients whose abscess cavities were packed after drainage compared with those who were not packed. ³⁸ The largest study to address this question was the Packing of Perianal Abscess Cavities (PPAC2) trial, a multicenter randomized controlled trial that enrolled 433 participants across 50 sites. It showed that those allocated to packing reported higher pain scores and there were no differences between the two groups regarding recurrent abscess or development of fistula. ³⁹

Needle Aspiration

In a randomized multicenter trial of 98 adults with a perianal abscess, patients were allocated to either needle aspiration and 1 week of clindamycin or traditional incision and drainage. The recurrence rate with needle aspiration was 41% compared with 15% with standard treatment. ⁴⁰ Therefore, needle aspiration cannot be recommended as an alternative to surgical drainage.

Use of Drainage Catheters

Small drainage catheters, such as a de Pezzer mushroom catheter or a Malecot drain, typically 12 to 14 Fr, can be used as an adjunct in the treatment of anorectal abscesses ( Fig. 3 ). Several studies compared drainage catheters to conventional packing and found that patients who had a mushroom catheter placed after incision and drainage had higher satisfaction scores and no difference in recurrence or fistula formation compared with patients who underwent traditional packing. ⁴¹

Fig. 3.

Anorectal abscess following drainage and placement of a Malecot catheter.

An approach that is utilized more frequently in the pediatric population is the use of a series of small counter incisions to drain the abscess cavity and using Penrose drains or vessel loops through the incisions to allow for continued decompression and drainage. This approach was studied in 91 adult patients and it showed no increase in recurrence or complications compared with traditional unroofing of the abscess. ⁴² In a series of 140 patients treated only with a stab incision and placement of a de Pezzer catheter for 2 weeks, 53 patients returned to normal activities or work within 5 days, and only 3 required further treatment for an inadequately drained abscess. ⁴³ It is important to note that if alternative techniques are utilized, one must ensure complete drainage of the abscess cavity as readmissions and reoperations are most commonly due to inadequate index drainage. ⁴⁴

Use of Antibiotics

After adequate drainage, the use of antibiotics is suggested in certain clinical scenarios, specifically in the setting of sepsis, significant surrounding cellulitis, or in immunocompromised patients. ^{13 45} According to the American Heart Association guidelines, antibiotics are also recommended before incision and drainage in patients with previous bacterial endocarditis, prosthetic valves, congenital heart disease, and heart transplant recipients with valve pathology. ⁴⁶

The role of antibiotics in fistula prevention is controversial. The Italian Society of Colorectal Surgery guidelines suggest that the use of antibiotics can decrease the rate of fistula formation following abscess drainage. ⁴⁷ They cite a 2017 randomized single-blinded study of 307 patients who underwent incision and drainage of a perianal abscess and were randomly allocated to received 7 days of oral metronidazole and ciprofloxacin or standard care after discharge. This study found a protective effect of postoperative antibiotics against the formation of a fistula at 3 months. ⁴⁸ In contrast, a 2011 randomized double-blinded placebo-controlled multicenter trial of 151 patients showed that antibiotic treatment following drainage of an anorectal abscess had no protective effect on the risk of fistula formation at 1 year. ⁴⁹ A recent meta-analysis of six studies found that the fistula rate in subjects receiving antibiotics was 16 versus 24% in those not receiving postoperative antibiotics and that this was a statistically significant difference. ⁵⁰ The Antibiotic Treatment foLlowing surgical drAinage of perianal abScess (ATLAS) trial is an ongoing multicenter, double-blind, placebo-controlled trial that is randomizing 298 patients to receive a 7-day course of ciprofloxacin and metronidazole or standard care following drainage to establish whether the addition of postdrainage antibiotics results in decrease rates of fistula formation at 1 year. ⁵¹ The most recent guidelines from the American Society of Colon and Rectal Surgeons do not make a recommendation for or against the use of antibiotics for the prevention of subsequent anal fistulas. ¹⁹

Obtaining Culture Data

Most perianal abscesses are polymicrobial, with Escherichia coli being the most common organism followed by Bacteroides spp. ^{10 52} Many studies have shown that culture results do not predict abscess recurrence, fistula formation, or lead to a change in clinical management, and do not support routine culturing of anorectal abscesses. ^{53 54 55}

There are a few scenarios where culture data may be helpful. Methicillin-resistant Staphylococcus aureus (MRSA) was found in 34.8% of cases where culture data were obtained, and guidelines recommend treatment of this organism in the setting of sepsis or immunocompromise. ^{56 57} One study of 361 patients found that two-thirds of patients with available culture data grew drug-resistant bacteria. This was associated with higher rates of re-debridement and systemic infection, suggesting a role for targeted antimicrobials in this setting. ⁵⁸

Special Considerations in Neutropenia

Rates of anorectal infections in neutropenic patients are reported to be upward of 10% and can be a major source of mortality. ^{59 60} Because neutrophils are critical in the formation of an abscess, these patients often present without the typical signs of erythema and fluctuance and often do not have a target for surgical drainage. ⁶¹ Rectal examinations should be undertaken with caution and only if necessary. Broad-spectrum antibiotics alone can be utilized, especially if there is no defined abscess. ⁶² If a clear collection is seen on examination or cross-sectional imaging, consideration should be given to incision and drainage, especially if antibiotics alone are failing to resolve the process. Also, it is not uncommon for these patients to develop abscesses as their cell counts recover. As such, surgical intervention may be required at a later date. ⁶²

Recurrence and Rate of Fistula Formation

Inadequate drainage is the primary reason for patients to have to undergo early repeat drainage. ⁶³ In a retrospective case study of 500 consecutive patients, 7.6% required reoperation within 10 days of the original procedure. The main factors that led to reoperation were inadequate incision size, premature closure of the skin edges over the drain cavity, and completely missed abscesses. Horseshoe abscesses were associated with a particularly high rate of operative failures (50%). ⁴⁴ Obesity was also found to be an independent predictor of abscess recurrence. ⁶⁴ In all studies, the rate of fistula formation is approximately 20 to 50%, with most fistulas developing within the first few months of the initial incision and drainage procedure. ^{49 65 66} Meanwhile, the median time from abscess drainage to diagnosis of a fistula was 7.0 months. ¹⁸

Role of Fistulotomy during Index Procedure

Performing a primary fistulotomy at the index drainage procedure remains controversial. Some argue that no attempt should be made to look for a fistula as the surrounding inflammation makes it difficult to identify a track and increases the risk of creating a false passage. ^{67 68} Additionally, many abscess drainage procedures are done by trainees and surgeons without specific colorectal training, and so one viewpoint is that they should be kept as simple as possible to avoid iatrogenic injury. ^{69 70} In the largest series of 1,000 patients where primary fistulotomy was performed, 7% of patients endorsed short-term changes in anal function and 3.2% at 2 years. ⁷¹ Smaller studies also support this finding. In a review of 51 cases of patients with an anorectal abscess who had a fistula identified and underwent primary fistulotomy, no recurrences were observed, but 17% had difficulty controlling flatus postoperatively. At 3 years, 6% still had impaired flatus control, while 4% reported fecal leakage. ⁷² Another study showed that patients treated with primary fistulotomy subsequently had a severe incontinence rate of 4%. This was attributed to the laying open of a fistula that was mistakenly classified as a low transsphincteric fistula. The group recommended that if a fistula was identified, then a loose seton can be placed to reduce the risk of sphincter damage. ⁷³

Others argue that primary fistulotomy is an appropriate treatment if an internal opening is identified and the fistula is suitable for division, as it leads to a very low (1.8% in one series) recurrence rate. ⁷⁴ With proper patient selection, several small, randomized trials have shown positive results with primary fistulotomy. Ho et al conducted a randomized controlled trial of only those who presented with a perianal abscess with the argument that a fistulotomy in these patients would still fully preserve the external sphincter. There were 28 individuals in the incision and drainage group and 24 in the fistulotomy group. At the 1-year follow-up, 25% of the incision and drainage group had developed a fistula, while none of the patients in the fistulotomy group developed a fistula. There were no changes in continence between the two groups. Interestingly, this study included anal manometry performed just before, 6 weeks, and 3 months after surgery, and these anal pressure readings were not significantly different between the two groups. ⁷⁵ Tang et al ⁷⁶ randomized 45 patients who presented with a perianal abscess and were found to have either a low intersphincteric or a low transsphincteric fistula. At the 1-year follow-up, one patient who had undergone fistulotomy had an anal stricture, and none had symptoms of fecal incontinence. A larger randomized study of 200 patients published in 2003 showed similar findings. Again, fistulotomy was performed only for subcutaneous or low trans- or intersphincteric fistulas. ⁷⁷ A Cochrane review published in 2010 showed that fistula surgery with concurrent abscess drainage reduced recurrence and the need for repeat surgery with no statistically significant worsening in continence. ⁷⁸

Despite these favorable outcomes following primary fistulotomy, an anal fistula occurs, at most, 50% of the time following drainage of an anorectal abscess. This means that half of all patients never have an additional problem following their primary drainage procedure. As such, pursuing a fistulotomy exposes them to a real risk of change in their continence that they otherwise would not have faced had the clinician simply performed an incision and drainage.

Role of Ligation of Intersphincteric Fistula Tract during Index Procedure

Ligation of intersphincteric fistula tract (LIFT) is a procedure typically performed in the treatment of an established transsphincteric anal fistula. ⁷⁹ A recent retrospective study examined the feasibility and outcomes of this procedure in 86 patients who presented with acute anorectal cryptoglandular abscesses. ⁸⁰ In addition to drainage of the abscess, all patients underwent an intersphincteric exploration. For the 66 patients where an identifiable tract was found, a LIFT was performed. An attempted closure of the internal opening was performed in the remaining 20 patients. Nonhealing was defined as abscess recurrence, presence of an external opening, or persistent intermittent discharge at 90 days. Nonhealing occurred in 14% of patients who underwent successful LIFT and in 30% of patients who only had closure of the internal opening. Of note, all patients received postoperative antibiotics for 1 week and there were no reports of fecal incontinence. Clinical practice guidelines from the American Society of Colon and Rectal Surgeons discuss the role of a draining seton prior to performing a LIFT, although this has not been shown to improve the success of a LIFT repair.

Conclusion

The optimal management of most superficial perianal abscesses involves bedside incision and drainage. Patients with larger abscesses, unclear external signs of an abscess, presence of sepsis, or on blood thinners should undergo incision and drainage in the operating room. The incision should be performed as close to the anal canal as possible to minimize the length of a possible subsequent fistula tract. Antibiotics are reserved for those with significant surrounding cellulitis or immunosuppression including the presence of diabetes. The safety of primary fistulotomy is debatable and we recommend against it.