9 Anorectal Abscess and Fistula-In-Ano
Abstract
A majority of abscesses are of nonspecific cryptoglandular origin but may also be due to a variety of processes, which are primarily inflammatory or traumatic. A fistula typically begins in the middle of the anal canal at the level of the crypts and extends downward to the anal verge. Penetration through the external sphincter will lead to an abscess into the ischioanal fossa, while upward extension by way of the intersphincteric plane will course along the rectal wall or penetrate beyond the rectum. High upward extension can enter the true pelvis. The most common presenting symptoms for anorectal abscess are pain and swelling. Presentation of fistula-in-ano may be subtle, but most patients will give a history of prior abscess or anorectal procedures. This chapter focuses on etiology and pathology, avenues of extension, diagnosis, investigations, treatment, and complications related to anorectal abscess and fistula-in-ano.
9.1 Anatomy
A thorough understanding of pelvic floor anatomy is a critical prerequisite in the evaluation and management of anorectal abscess and fistula. A comprehensive discussion of the relevant anatomy is provided elsewhere in this text (Chapter 1). For our purposes, the anal sphincter complex and its immediate environment can be thought of as two funnels of muscle surrounding the distal rectum and anal lumen. The internal anal sphincter is the thickened and rounded distal extent of the inner circular smooth muscle layer of the rectum. It terminates just distal to the dentate line and remains in a constant tonic contraction, contributing to continence. The internal sphincter is nested within a broad funnel of striated muscle consisting of the external anal sphincter, levator ani, and puborectalis. The external anal sphincter extends slightly more caudad than the internal sphincter, and the intersphincteric groove between them can be visualized and palpated. At the level of the dentate line, a variable number of anal glands empty into crypts (▶ Fig. 9.1).
9.2 Etiology and Pathogenesis
A majority of abscesses are of nonspecific cryptoglandular origin. The remainder are due to a variety of processes, which are primarily inflammatory or traumatic. The specific processes that can lead to anorectal abscesses are as follows:
Infectious/inflammatory
Cryptoglandular abscess
Inflammatory bowel disease
Diverticulitis
Pelvic abscess
Tuberculosis
Actinomycosis
Lymphogranuloma venereum
Malignancy
Carcinoma
Lymphoma
Leukemia
Trauma
Impalement
Penetrating trauma
Episiotomy
Hemorrhoidectomy
Prostatectomy
Enemas
Radiation
The cryptoglandular origin theory was first described by Eisenhammer in 1958 and further supported by Parks’ anatomic studies in 1961. 1 , 2 These anatomic studies demonstrated that anal glands provide a passage from crypts in the lumen of the bowel, through the internal sphincter, and into the intersphincteric space. Parks wrote that bacterial infection becomes established in glands that are either cystic or whose lumen is obstructed. While some glands may become obstructed due to stool, foreign bodies, trauma, or inflammation, there is no clear source of obstruction for most patients. Once infection is established, the epithelialized gland remains patent to the bowel lumen, while pus drains to the skin via any path of least resistance, thereby forming a fistula.
9.3 Avenues of Extension
A fistula typically begins in the middle of the anal canal at the level of the crypts and then most often extends downward in the intersphincteric plane to the anal verge. Penetration through the external sphincter will lead an abscess into the ischioanal fossa, while upward extension by way of the intersphincteric plane will course along the rectal wall or penetrate beyond the rectum. High upward extension can enter the true pelvis (▶ Fig. 9.2).
Pus may also track circumferentially around the anus in one of three tissue planes. The most common of these is known as a posterior horseshoe and begins with an abscess in the posterior midline that penetrates the sphincter complex into the deep postanal space and tracks into the bilateral ischioanal fossae. Circumferential spread may also occur in the intersphincteric plane or the pararectal tissue above the levator muscles.
9.4 Diagnosis
9.4.1 History
The most common presenting symptoms for anorectal abscess are pain and swelling. 3 While the patient will most often complain of perianal pain, the clinician should also be wary of supralevator abscess in the patient who complains of generalized pelvic or gluteal pain. Other symptoms of abscess include swelling, purulent drainage, odor, difficulty urinating, bleeding, and fevers. Patients who are immunocompromised by disease or medication may have subtle presentations with fevers and malaise without pain or swelling. Patients who give a history of prior anorectal abscess, inflammatory bowel disease, or anal trauma often have more complex abscess, and early consideration should be given to adjunct means of evaluation including examination under anesthesia and imaging.
Presentation of fistula-in-ano may be subtle, but most patients will give a history of prior abscess or anorectal procedures. Symptoms of fistula include drainage, bleeding, pain, and swelling. The fistula tract may intermittently occlude, leading to recurrent abscess formation.
9.4.2 Physical Examination
Findings on physical examination of the patient with acute abscess will vary depending on the location of the abscess. A superficial perianal or ischioanal abscess may be characterized by erythema, induration, fluctuance, tenderness, and occasionally purulent drainage. Evaluation of an intersphincteric abscess deserves special mention as the patient may present without erythema, induration, or swelling of the perineum. The patient will complain of severe perianal pain and tenderness with attempted rectal examination but no external signs of inflammation. Supralevator abscesses may also present with pain and tenderness without physical examination findings. Whereas patients with an intersphincteric abscess may not tolerate digital rectal examination due to discomfort, patients with a supralevator abscess, beyond the sphincter complex, may allow passage of an examining finger to elicit tenderness more cephalad in the pelvis.
The most common physical examination finding of a fistula-in-ano without acute suppuration is the external opening. This is characterized as a small elevated sinus with granulation tissue. With compression, this may express serous or purulent fluid. Palpation may reveal the firm cord of the fistula tract within the perianal tissue. Anoscopy on the awake patient rarely reveals the internal fistula opening.
Goodsall’s rule is an often-cited method for identifying a fistula’s internal opening based on the location of its external opening (▶ Fig. 9.3). According to the first published account of Goodsall’s observation in 1887, “For fistulae having their external orifices situate behind a plane passing transversely through the center of the anus usually have their internal aperture in the middle line dorsally, while those with their external orifice in front of this plane generally terminate in an internal opening immediately opposite, thus forming a simple, straight, complete fistula.” 4 Several studies have cast doubt on the utility of Goodsall’s rule. In one, Goodsall’s rule accurately predicted the site of the internal opening in approximately 59% of fistula tracts. It was most accurate at identifying anterior openings (72%) compared to posterior openings (41%). 5 Interestingly, an earlier but larger study of 216 patients suggested that Goodsall’s rule was 90% accurate for identifying posterior fistulas that track to the midline but only 49% accurate for identifying the internal opening of an anterior fistula. 6
9.4.3 Anoscopy and Proctoscopy
Examination under anesthesia with a patient comfortable and optimally positioned, and with ideal lighting and necessary equipment, provides an opportunity for the most complete evaluation of anorectal abscess and fistula. These procedures are easily performed in the prone position, although occasionally examination in the lithotomy or lateral decubitus position is necessary. General or local anesthesia sufficient to allow an unhurried examination should be administered. Anoscopy may allow identification of a crypt associated with an internal opening draining pus. Intersphincteric abscesses may be identified by palpation or passage of an 18-gauge finder needle.
Evaluation of fistula-in-ano requires identification of the internal opening whenever possible. Inserting a fistula probe into a chronic, mature fistula tract is an acceptable method for delineating the path of the fistula but must be done with caution to avoid formation of false channels. With an examining finger in the anus, the probe is gently advanced as long as resistance is not met. If the probe does not pass easily, dilute methylene blue or hydrogen peroxide can be instilled in the external opening to highlight an internal opening. Another method is to enlarge the external opening and dissect following the granulation tissue of the fistula tract, again taking care not to create false passages.
Anoscopy and proctoscopy or flexible sigmoidoscopy are also important if the abscess or fistula could be attributed to an underlying disease such as inflammatory bowel disease, malignancy, ischemia, or foreign body. Every patient undergoing treatment for fistula-in-ano should, at a minimum, be evaluated for proctitis.
9.4.4 Investigation
Magnetic Resonance Imaging
The ability of MRI to noninvasively evaluate distorted anatomy in patients with complex fistulas or prior perineal surgery makes it a useful tool in the management of selected patients with anorectal abscess or fistula (▶ Fig. 9.4, ▶ Fig. 9.5, ▶ Fig. 9.6). Initial interest in endoanal MRI coil has been tempered by its lack of availability and patient discomfort. The majority of studies therefore rely on body coil imaging. Pelvic MRI is described as the gold standard imaging technique for perianal fistulas associated with the complex anatomic disruption often found with Crohn’s disease. 7
An early study of MRI for fistula-in-ano using surgical findings as the gold standard in 35 patients demonstrated concordance of 86% for the presence and course of the primary tract, 93% for the presence and site of secondary extensions, and 80% for the position of the internal opening. This included two patients in whom an initial examination under anesthesia was unable to locate fistula tracts previously identified by blinded MRI but then required repeat surgical drainage after developing perianal sepsis at the site indicated by MRI. 8 Recent studies using more powerful MRI magnets have supported these findings and have particularly confirmed the ability of MRI to reliably identify secondary tracts. Sensitivity for extension or secondary tracts was 80 to 94%, and specificity was 94 to 100%. 9 , 10 Meta-analysis of four studies reported combined sensitivity and specificity of MRI for overall fistula detection of 87 and 69%. This was compared to sensitivity and specificity of 87 and 43% for endoanal ultrasound. 11
One study by Buchanan et al compared clinical examination, endoanal ultrasonography, and MRI to an outcome-derived reference standard in 104 patients. In this study, clinical examination, ultrasonography, and MRI were able to correctly classify 61, 81, and 91% of possible fistula tracts and 36, 70, and 88% of abscesses, respectively. 12
Fistulography
Fistulography involves cannulation of the external fistula opening and injection of water-soluble contrast material while multiple fluoroscopic images are obtained. Its utility is limited as there is poor visualization of soft-tissue anatomy including the absence of landmarks to localize internal openings.
Recent global consensus guidelines for fistula management in Crohn’s disease state that fistulography and CT are outdated modalities as they provide poor anatomic relationship between the fistula and the pelvic floor muscles. 7 There may be a role for fistulography in certain exceptionally complex fistulas as it may change operative strategy in some. 13 A more recent study utilizing radiopaque markers to delineate anatomy for fistulography found this modality to be 74% accurate in identifying the presence or absence of an internal opening. 14 Other authors have shown that the rate of fully correlating fistulography with operative findings is only 16%. 15
Ultrasonography
Endoanal ultrasound is a useful alternative to MRI for evaluation of fistulas as it may be more readily available and has similar accuracy. Its usefulness is limited by smaller field of view and by its inability to adequately visualize abscesses in the supralevator or ischioanal spaces. Hydrogen peroxide and saline fistulography are useful adjuncts to improve visualization. 7 Using definitive surgical findings as the reference standard, Toyonaga et al found endoanal ultrasound to be 88% accurate at detecting a primary fistula tract, 85.7% accurate at describing horseshoe extension, and 85.5% accurate for localizing the primary fistula opening in 401 patients. 16 This study selectively used hydrogen peroxide injected into the fistula opening to enhance visualization (4.9%). Other studies have shown similar results using peroxide-enhanced ultrasound. 17 Directly comparing peroxide-enhanced ultrasound and MRI has demonstrated close correlation between the findings of these two modalities. In one study, the agreement between ultrasound and MRI was 90% for classification of the primary tract, 71% for secondary tracts, and 90% for location of internal openings. 18
Computed Tomography
The utilization of CT imaging for evaluation of anorectal fistula has been limited due to the inability to provide sufficient soft-tissue resolution around the anal sphincter. The entry of multiplanar CT imaging systems with improved resolution has rekindled interest in CT fistulography as a less expensive and more readily available alternative to MRI. One recent study compared CT fistulography and MRI to operative findings in 41 patients. CT was able to correctly predict fistula classification in 73% of patients, number of secondary extensions in 85% of patients, and quadrant of the internal opening in 68.2% of patients compared to 92.7, 87.8, and 85.3%, respectively, for MRI. These authors concluded that MRI is superior to CT fistulography in most cases, but CT fistulography could be considered the first choice in uncomplicated cases. 19 As uncomplicated fistulas rarely require radiologic evaluation, this study provides little support for invasive imaging involving ionizing radiation for simple fistulas.
Endoanal ultrasonography and pelvic MRI may add useful information in the evaluation of anorectal abscess and fistula in selected, complex patients. Fistulography is not widely practiced and few published studies support its use. The widespread availability of CT scan and the speed with which it can be obtained may make it useful in the evaluation of an acute anorectal abscess. However, MRI and ultrasonography are the preferred modality for evaluation of complex fistulas or small abscess due to superior resolution of sphincter complex anatomy without exposure to radiation.
9.5 Anorectal Abscess
9.5.1 Microbiology
Bacterial isolates from anorectal abscesses typically represent a mix of aerobic and anaerobic normal bowel flora and are rarely resistant to common antibiotics. Organisms typically isolated include Escherichia coli, Bacteroides fragilis, Klebsiella pneumoniae, coagulase-negative Staphylococci, and Enterococcus species. 20 , 21 , 22 There may be an association between bowel-related organisms cultured from acute abscess and fistula formation. One study reported that 54% of patients whose anorectal abscess cultures grew bowel-derived organisms developed fistulas, compared to 0% of 34 patients whose cultures grew skin-derived organisms, and therefore the authors advocate for a second surgical evaluation to search for fistula in those patients whose anal abscess grows bowel-derived organisms. 23 Most conclude that aggressive pursuit of a fistula based on microbiologic data is unnecessary, as the majority of patients with bowel-derived organisms will not develop a fistula. 24 They suggest that a delayed fistula procedure be performed later if necessary. Microbiologic evaluation of anorectal abscess has been largely absent from recent studies as it is felt to be a somewhat sensitive but not specific means of evaluating for fistulas and inferior to careful anatomic evaluation. 25 One study from 2014 found that 98% of bacteria isolated from anorectal abscesses were sensitive to common oral antibiotics, while 2/172 cultures grew methicillin-resistant Staphylococcus aureus. 26 Our practice is to limit the collection of anorectal abscess cultures to those patients who may harbor multidrug-resistant bacteria due to extensive prior antibiotic exposure.
9.5.2 Incidence and Classification
The exact incidence of anorectal abscess is unknown as no national registries exist for outpatient or in-office procedures. Nelson performed a fascinating extrapolation of abscess incidence based on historical data on fistula incidence and an assumption that a defined percentage of abscesses will result in fistulas. His analysis suggests that in the United States the incidence of anorectal abscess is 68,000 to 96,000 cases per year. 27 Several large series have shown a higher incidence in males than females, with ratios ranging from 2:1 to 5:1. 3 , 28 , 29 , 30
Anorectal abscesses are classified based on anatomic location within potential spaces: perianal, ischioanal, intersphincteric, and supralevator (▶ Fig. 9.7). Posterior abscesses that enter the deep postanal space by penetrating the posterior midline conjoint longitudinal muscle and extend to one or both ischioanal spaces are described as a horseshoe abscess. These are often discussed as a separate classification of abscess as their management can be complex. In a review of 1,023 patients with anorectal abscess at Cook County Hospital, Ramanujam et al found the following frequency of anatomic distribution: perianal, 42.7%; ischioanal, 22.8%; intersphincteric, 21.4%; supralevator, 7.3%; submucous (high intermuscular), 5.8%. 3
9.5.3 Treatment
Role of Antibiotics
The use of antibiotics in anorectal abscess should be limited to a few specific clinical scenarios as there is no evidence that their use improves healing time or decreases complications. Seow-En and Ngu retrospectively reviewed 172 patients who underwent drainage of anorectal abscess; 63% were prescribed postoperative antibiotics. Despite the fact that the mean abscess size was significantly larger in the group that received antibiotics, there was no difference in abscess recurrence. 26 A randomized, double-blinded, placebo-controlled, multicenter trial of antibiotics after anorectal abscess drainage failed to reveal an improvement in rate of fistula formation in patients randomized to postoperative antibiotics. Rate of fistula formation was lower in patients randomized to placebo, when compared to those given antibiotics (22.4 vs. 37.3%). 31
The Clinical Practice Guidelines from the American Society of Colon and Rectal Surgeons state: “… the addition of antibiotics to routine incision and drainage of uncomplicated anorectal abscess does not improve healing time or reduce recurrences, and it is therefore not indicated.” However, there are specific circumstances for which antibiotics might complement incision and drainage: cellulitis, immunosuppression, concomitant systemic illness, prosthetic heart valves, prior bacterial endocarditis, congenital heart disease, or heart transplant with valvular disease. 32
Perianal Abscess
The majority of simple perianal abscesses can be drained under local anesthesia in the office or when seen in the emergency department. Generous injection of local anesthetic with epinephrine at and around the point of maximum fluctuance allows complete drainage and can improve hemostasis. A cruciate incision may be created and the skin edges excised (▶ Fig. 9.8) or an ellipse of skin may be excised. Some overlying skin must be removed, as simple stab incision may allow pus to reaccumulate as the skin edges heal closed. We typically place packing if hemostasis is needed, or place a small bulb-tipped catheter if the cavity is large.
In general, there is no need for prolonged packing or repeated dressing changes of the abscess cavity as this is uncomfortable for the patient and provides no long-term benefits. Two small randomized studies have compared packing versus not packing these wounds. In each study, after incision and drainage of perianal abscess, a hemostatic packing was placed. Patients were then randomized to daily nursing packing changes or no packing. One study evaluated 43 patients and found no difference in time to healing, recurrent abscess formation, rate of fistula formation, pain scores, or length of stay. 33 A more recent study evaluated 14 patients and found faster mean time to healing in the nonpacking group (19.5 vs. 26.8 days, p = 0.047) along with lower pain scores at 2 weeks in the nonpacking group. There was no difference in recurrence rates, and a single fistula was reported in the packing group. 34
Ischioanal Abscess
Ischioanal abscesses can typically be managed using the same general principles as perianal abscesses. A smaller abscess in a compliant patient can often be incised and drained under local anesthesia. A larger abscess may require drainage under anesthesia to allow complete evaluation for loculations and counterincisions as needed to completely drain the cavity. Again, excision of a sufficient amount of skin to allow complete drainage is advised and packing is unnecessary.
We will often supplement incision and drainage of a large ischioanal abscess with drainage catheter insertion into the abscess cavity. In the initial description of this technique by Isbister, 91 patients were managed with simple Pezzer catheter insertion. The procedure was generally well tolerated and 22/91 (24.2%) patients developed fistulas. 29 In a large anorectal abscess, it has been our practice to secure a Pezzer catheter in the abscess cavity for several weeks as the cavity collapses. The catheter is then removed in the office and ongoing drainage or recurrent suppuration is evaluated for fistula.
Intersphincteric Abscess
As previously discussed, the diagnosis of an intersphincteric abscess often requires a high degree of clinical suspicion as patients may present with pain but no external signs of inflammation. Focal tenderness and fluctuance on digital rectal examination can direct the health care provider to the abscess. Incision and drainage of an intersphincteric abscess will often require general or regional anesthesia for patient comfort and to allow adequate exposure. Treatment consists of drainage of the abscess internally, including division of internal anal sphincter fibers overlying the abscess cavity.
Supralevator Abscess
Like intersphincteric abscess, supralevator abscess can often be difficult to diagnose as findings on physical examination can be few and subtle. The patient may complain of severe perianal pain with no external evidence of inflammation. The source of a supralevator abscess determines its proper management and should therefore be determined whenever possible before treatment begins. The three possible origins of a supralevator abscess are: cephalad extension of an intersphincteric abscess, cephalad extension of an ischioanal abscess, or caudad extension of a pelvic abscess caused by diverticulitis, appendicitis, Crohn’s disease, or other source. A combination of evaluation under anesthesia and radiographic examination may be necessary to determine the abscess origin. Both CT and MRI have proven valuable in the evaluation of supralevator abscess. 35 , 36
If the supralevator abscess is a result of intersphincteric or ischioanal abscess, then it should be drained as would be appropriate for its progenitor abscess. That is, a supralevator abscess of intersphincteric origin should be drained internally by dividing internal sphincter overlying the abscess cavity. Drainage of an intersphincteric-derived supralevator abscess via the ischioanal fossa may produce a suprasphincteric fistula. A supralevator abscess of ischioanal origin should be drained through the ischioanal fossa and not via internal drainage with division of internal sphincter fibers. Internal drainage of an ischioanal-derived supralevator abscess may produce an extrasphincteric fistula (▶ Fig. 9.9). Management of the third type of supralevator abscess, one created from downward extension of a pelvic process, is determined by the original disease process. Typically, image-guided catheter drainage of the pelvic abscess provides the least morbid means of treatment. However, surgical drainage through the rectal lumen, through the ischioanal fossa, or through the abdominal wall may be necessary.
Horseshoe Abscess
Lateral extension of an anorectal abscess is commonly described as a horseshoe abscess. This description may be applied to an intersphincteric abscess that has simply extended from the anterior or posterior midline to involve the lateral potential intersphincteric space. This is managed like other intersphincteric abscesses by dividing internal sphincter and allowing internal drainage. The more complex horseshoe abscess is one that has penetrated the conjoint longitudinal muscle at the posterior midline, involved the deep postanal space, and may have spread to one or both ischioanal fossae. The location of the deep postanal space, deep to the external sphincter but superficial to the levator muscles, is important to appreciate as it is relevant to both the formation of these abscesses and their management.
Management of the deep postanal space horseshoe abscess must focus on adequate drainage of this space and was famously described by Hanley. Hanley’s original technique includes performing a fistulotomy by inserting a fistula probe in the posterior midline primary fistula opening and dividing the overlying superficial external sphincter in the posterior midline between the primary fistula opening and the coccyx. This allows access to the deep postanal space. Counterincisions may then be created in the ischioanal fossae on either side to allow complete drainage of the lateral extensions of the horseshoe abscess (▶ Fig. 9.10). 37 Instead of completely laying open the posterior midline external sphincter, Hamilton described a series of 65 patients in whom the deep postanal space was accessed by making a skin incision in the posterior midline and spreading the sphincter fibers. To avoid incontinence, some patients with deep or complex abscesses underwent staged procedures with seton placement. Follow-up data were available for 57 of these patients and 4 recurrences were reported. 38 In later years, Hanley described this modification of his lay-open procedure by placing a cutting seton in the primary fistula tract instead of performing an immediate fistulotomy (▶ Fig. 9.11). 39 As cutting setons have largely fallen from favor, our practice has been to place a loose seton and defer management of the fistula to one of the myriad of less destructive options that have proliferated in recent years.
Recurrent Abscess
Diagnostic evaluation and management of a recurrent abscess can be challenging. Patients may have undergone multiple drainage procedures before referral to a specialist and the resultant anatomic disruptions can complicate evaluation. Early recurrence should raise concern for inadequate drainage. In a study of 500 patients who underwent incision and drainage of anorectal abscess, Onaca et al reported 48 early reoperations within 10 days of the index procedure. As several patients required more than one reoperation, the overall reoperation rate per procedure was 7.6%. Early reoperation was due to inadequate drainage in 23/48 (48%) or missed loculations in 15/48 (32%). Reoperation rate was highest for supralevator abscesses (33%) and low for perianal, ischiorectal, and intersphincteric abscesses (6.9, 7.8, and 7.3%, respectively). Only age older than 21 was a significant risk factor for reoperation, while diabetes, steroids, time from symptom onset to surgery, primary or recurrent abscess, associated fistula, and Crohn’s disease were not significant risk factors. Interestingly, there was no difference in the reoperation rate for procedures by junior residents compared to those performed by senior residents or attending surgeons. 40
A more recent study of recurrent abscesses reported the need for early reoperation in 2.4% and a 36.1% long-term recurrence at a mean of 20 months’ follow-up. Fistula was diagnosed in 20% of patients with recurrence. Shorter time from onset of symptoms to drainage was the only factor associated with recurrence with a hazard ratio of 0.4 (95% confidence interval [CI] = 0.23–0.68, p = 0.001). Gender, age, BMI (body mass index), method of anesthesia, abscess location, anatomic classification, drain use, and diabetes were not significant. 41 Ramanujam et al reported 25 recurrences out of 663 patients who underwent incision and drainage of anorectal abscess, and the majority (22/25) were found to have a fistula. 3
Primary Fistulotomy
Fistulotomy when draining the index abscess is controversial. Proponents of the search for and destruction of a fistula tract at the time of abscess drainage cite a decrease in the rate of recurrence and the need for future surgery. Opponents argue that the majority of anorectal abscesses treated with incision and drainage will not recur or form a fistula, and the search for a fistula converts a simple drainage procedure to an unnecessarily complex procedure. In addition, the presence of acute inflammation may increase incidence of false passages in the search for a fistula, or increased risk of incontinence upon division of excessive sphincter muscle.
Large retrospective series have shown that fistula develops in approximately one-third of patients who undergo incision and drainage of anorectal abscess and another 10% suffer from recurrent abscess. 24 , 28 Results of these selected series are shown in ▶ Table 9.1. These authors have concluded that primary fistulotomy at the time of abscess drainage is unnecessary as less than 50% of patients who undergo incision and drainage go on to require further procedures. The remainder, in whom incision and drainage is sufficient treatment, derive no benefit from primary fistulotomy and risk complications.
However, a recent meta-analysis evaluated 479 patients from six studies randomized to drainage or drainage with primary fistulotomy. The majority of these trials excluded patients with prior fistula surgery, Crohn’s disease, suprasphincteric fistulas, or extrasphincteric fistulas. Meta-analysis showed a significant reduction in recurrence after primary fistulotomy (risk ratio [RR] = 0.13, 95% CI = 0.07–0.24) and a trend toward increased incontinence at 1 year (RR = 3.06, 95% CI = 0.7–13.45). Because of significant heterogeneity in the incontinence data, these authors conclude that patients overall benefit from decreased recurrence and the risk of harm is minimal when low fistulas are treated with primary fistulotomy. 42 With these results in mind, we suggest that there is a role for primary fistulotomy in properly selected patients, such as those who have a clearly identifiable internal opening of a low intersphincteric or transsphincteric fistula without significant comorbid conditions or prior anorectal surgery.
9.5.4 Postoperative Care
After abscess drainage, the patient can typically be discharged home directly from the office, emergency room, or postoperative recovery room. The patient should be counseled on the risk of abscess recurrence or fistula formation and given a reasonable timeline for wound healing. The risk of recurrence after drainage of an abscess is summarized in ▶ Table 9.1 .
Nonhealing wounds or persistent drainage should prompt evaluation for a fistula. Sitz baths, stool softeners, fiber preparations, and narcotic or nonnarcotic analgesia should be recommended as appropriate. As previously discussed, prolonged wound packing and antibiotics are rarely indicated.
9.6 Fistula-In-Ano
9.6.1 Incidence
As with anorectal abscess, fistula-in-ano is a condition more commonly identified in the male population. Male-to-female ratios are generally reported around 2:1. 43 , 44 , 45 Overall incidence has been estimated to range from 1.04 to 2.32 cases per 10,000 population/year based on large, anonymous European databases. 46
9.6.2 Indications for Operation
In general, the presence of a symptomatic fistula-in-ano is in itself sufficient indication for surgical repair as spontaneous healing is uncommon. Symptoms typically include persistent drainage, recurrent abscesses, or discomfort. Although the majority of fistulas are associated with prior cryptoglandular abscess, the surgeon should also be wary of rare causes of fistulas including Crohn’s disease, hidradenitis suppurativa, tuberculosis, or malignancy. Patients presenting with a draining anorectal sinus tract without a prior history of abscess should be carefully evaluated for these conditions before planning repair.
Relative contraindications to fistula repair include medical risk factors for general or regional anesthesia, fecal incontinence, and unmanaged inflammatory bowel disease. In each of these cases, the surgeon must weigh the risks of the procedure against the patient’s symptoms. Often, recurrent abscess due to persistent fistula can be managed with long-term seton placement until underlying disease can be controlled.
9.6.3 Principles of Treatment
The principles of fistula surgery are individually simple but can present a complex challenge when considered as a whole. The goal is to heal the fistula in a reasonable amount of time, with the lowest possible recurrence rate, without disrupting continence. To accomplish these goals, several principles should be adhered to as described in ▶ Table 9.2.
9.6.4 Classification and Treatment
Multiple classification systems have been proposed to describe the anatomic relationships of fistulas-in-ano. The most beneficial system is one that allows communication between surgeons and helps inform treatment of the fistula. Therefore, precise description of the primary tract along with the fistula’s relationship to the muscular anatomy of the pelvic floor is necessary. The description by Parks et al is complex but provides this level of detail. 47 Four main classifications are determined by relationship between the sphincter muscle and the primary tract of the fistula: intersphincteric, transsphincteric, suprasphincteric, and extrasphincteric. Each of these can be modified based on secondary ramifications and extensions:
Intersphincteric fistulas
Simple low tract
High blind tract
High tract with rectal opening
Rectal opening without a perineal opening
Extrarectal extension
Secondary to pelvic disease
Transsphincteric fistulas
Uncomplicated
High blind tract
Suprasphincteric fistulas
Uncomplicated
High blind tract
Extrasphincteric fistulas
Secondary to anal fistula
Secondary to trauma
Secondary to anorectal disease
Caused by pelvic inflammation
What follows is a brief description of each type of fistula along with an illustration of the fistula. Each illustration contains the disease process on the left and the excisional treatment option on the right. These illustrations should be considered a general guide to management and the reader is reminded to refer to the following sections of this chapter for other management options that involve division of less sphincter muscle and may provide equal fistula healing rates. Surgical treatment of fistulas requiring resection of large amounts of sphincter muscle may result in permanent incontinence.
Intersphincteric Fistula
The intersphincteric fistula is the most common type of fistula and involves only the intersphincteric plane. It is often the precursor to other types of fistulas and has multiple possible extensions or secondary ramifications. A simple low tract intersphincteric fistula penetrates the internal sphincter and then passes to the anoderm. It can be treated with division of the distal internal sphincter muscle with little risk to continence (▶ Fig. 9.12). An intersphincteric fistula with a high blind tract adds an additional upward extension between the internal sphincter and the longitudinal muscle of the rectal wall. Identification and management of the high blind tract by division of more proximal sphincter muscle will decrease the chances of recurrence without adding significantly to the risk of incontinence (▶ Fig. 9.13). If this high blind tract penetrates back into the lower rectum, a second rectal opening may be produced. As with a high blind tract, this second opening must be identified and treated with division of further internal sphincter to prevent recurrence (▶ Fig. 9.14). If an intersphincteric abscess forms without draining to the perineum, it is considered an intersphincteric fistula without a perineal opening. These must also be treated with division of the lower portion of the internal sphincter to eliminate the causative crypt (▶ Fig. 9.15). Finally, an intersphincteric cryptoglandular abscess can extend into the pelvic cavity or a pelvic abscess caused by a primary intestinal process can spontaneously drain through the intersphincteric plane. Management is different in each of these cases as division of sphincter muscle plays no role in management of intersphincteric fistula caused by bowel perforation but is important in an intersphincteric fistula that extends proximally from a cryptoglandular abscess (▶ Fig. 9.16, ▶ Fig. 9.17).
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
