Beware the ischiorectal abscess

Abstract

INTRODUCTION

Ischiorectal abscesses have been shown to form sinuses with various deep structures but continuity with the spinal canal is extremely rare.

PRESENTATION OF CASE

A previously healthy sixty-five year old man presented emergently with rectal pain, weight loss and recurrent severe tension headaches. He had systemic sepsis and resultant coagulapathy (INR 3.4) which precluded investigation of neurological symptoms by lumbar puncture. MRI rectum demonstrated a well circumscribed fluid collection with direct connection to the spinal canal and containing meningeal tissue. It extended inferiorly to the right ischiorectal fossa and abutted the natal cleft. A radiological diagnosis of ischiorectal abscess which had become continuous with a previously existing anterior sacral myelomeningocoele (ASM) was made. He was treated with broad spectrum antibiotics and a neurosurgical opinion was sought. He remained clinically unwell (septic and coagulopathic) until the abscess fistulated through the perianal skin, draining pus mixed with clear fluid (likely CSF) at which point he improved systemically.

DISCUSSION

Few general surgeons would be faced with acute management of complicated ASM. Paucity of literature made application of evidence based medicine difficult. In fit healthy patients surgery is the mainstay of treatment as myelomengingoceles do not regress spontaneously. Conservative management is associated with up to 30% mortality (largely due to bacterial meningitis). The patient in this case was adamant that he did not consent to definitive surgical intervention.

CONCLUSION

This case highlights challenges encountered in the management of complicated ASM in a general hospital.

1. Case report

While ischiorectal abscess are frequently managed by surgeons providing emergency care, it is extremely rare for patients to present with associated meningeal symptoms. A majority of colorectal surgeons would never be faced with the management of anterior sacral myelomeningocoele in a lifetime practice. ASM results from herniation of meninges through a defect in the anterior aspect of the sacrum, forming a cyst-like structure filled with cerebrospinal fluid which is continuous with the subarachnoid space of the spinal cord.¹ The present case describes the unusual presentation and management dilemmas associated with a complicated ASM.

A previously healthy sixty-five year old male presented emergently to a general hospital with weight loss (5 kg), night sweats, recurrent severe tension headache, proctalgia, tenesmus and a sensation of rectal pressure over a one month period. His previous medical history included chronic constipation and intermittent urinary retention over the preceding five years. Colonoscopy to caecum (to investigate chronic constipation) had been performed one year earlier and subsequent barium enema six months prior to this presentation. Both were essentially normal.

On presentation to the emergency department he was systemically unwell with a temperature of 40.2 °C. He was hypotensive (80/40 mmHg) and had a sinus tachycardia (130 bpm). On digital rectal examination there was a tender boggy mass in the posterior aspect of the rectum. There was no obvious pointing inflammatory mass. Haematological and biochemical investigations confirmed systemic sepsis (White cell count 30 (normal value < 10) and C Reactive Protein 400 (normal value < 2). He had a resultant coagulopathy with an international normalised ratio (INR) of 3.4 which precluded investigation of neurological symptoms with lumbar puncture. Anaerobic and aerobic blood cultures grew streptococcus constellatus (an unusual pathogen often associated with abscess formation). He had a normal CT brain and subsequently had a CT abdomen and pelvis (Fig. 1). This demonstrated anterior displacement of the rectum by a cystic mass which appeared to be contiguous with the neural canal.

Fig. 1.

Saggital section of the CT abdomen and pelvis demonstrating anterior displacement of the rectum by a large cystic mass contiguous with the neural canal.

He then developed meningeal symptoms of headache, photophobia, and acute confusion with a swinging pyrexia. In view of systemic sepsis and impending septic shock with resultant coagulopathy he was admitted to the intensive care unit. An MRI pelvis was performed to establish a definitive diagnosis (Figs. 2 and 3a and b). It became clear that the ischiorectal abscess had ruptured into a pre-existing anterior sacral myelomeningocoele, giving rise to signs and symptoms of bacterial meningitis. Broad spectrum intravenous antibiotics (piperacillin and gentamicin) and antifungal therapy (fluconazole) were commenced. Coagulopathy was corrected in accordance with hospital haematology protocol. An urgent neurosurgical opinion was obtained from a specialist center by teleconference with relevant images reviewed. The neurosurgery team expressed significant concern about the potential risk of persistent cerebrospinal fluid leakage resulting from incision and drainage of the associated abscess. In view of this major risk, the neurosurgical team accepted transfer of the patient with a plan for incision and drainage of the abscess cavity in the specialist center. While awaiting transfer by ambulance, the abscess spontaneously fistulated through perianal skin (draining pus and cerebrospinal fluid) and the patient demonstrated a notable clinical improvement. A rectal swab grew enterococcus and coliforms sensitive to vancomycin. He had a sustained clinical improvement on antimicrobial and supportive management. He was offered definitive neurosurgical intervention (elective resection of the ASM and obliteration of the fistulous tract) but, following extensive discussion regarding potential risks and benefits the patient and his family opted for formation of defunctioning loop ileostomy and not to consent to neurosurgical management of the myelomeningocoele. Repeat MRI (Fig. 3a and b) demonstrated almost complete resolution of the ischiorectal abscess and an anterior sacral myelomeningocele (ASM) less than half the size of that on original presentation. He was discharged home and was asymptomatic on outpatient follow up six months later. He has decided that he will not consent to any future surgical intervention (including stoma closure).

Fig. 2.

MRI rectum demonstrating a well circumscribed fluid collection with direct connection to the spinal canal and containing meningeal tissue. It extends inferiorly to the right ischiorectal fossa and abutts the natal cleft.

Fig. 3.

(a) MRI (saggital section) demonstrating rectum sandwiched between the anterior sacral myelomeningocele and urinary bladder which is grossly distended in keeping with acute urinary retention. (b) Repeat MRI (4 weeks after initial presentation). Anterior sacral myelomeningocele has halved in size and only remnants of the ischiorectal abscess remain.

2. Discussion

There are five types of anorectal abscesses described: perianal (60%), ischiorectal (30%), intersphincteric (5%), supralevator (4%) and submucosal (1%). Anorectal abscesses usually arise from infection of the cryptoglandular epithelium along the anal canal. In the case of an ischiorectal abscess, as occurred in our patient, it is believed that pathogenic organisms breach the barrier normally provided by the internal anal sphincter and traverse the crypts of Morgagni, gaining access to intersphincteric and ischiorectal spaces. A collection of pus forms and an abscess cavity is created. Incision and drainage under general anaesthesia is the treatment of choice. The wound is left open to heal by secondary intention and interval colonoscopy is performed to outrule underlying causative processes such as inflammatory bowel disease. Urinary retention and neurological complaints would not generally be associated with an uncomplicated anorectal abscess and because of the unusual constellation of symptoms and coagulopathy, the admitting surgeon exercised caution in performing an emergency incision and drainage of the abscess. Computed tomography (CT) and magnetic resonance imaging (MRI) scans were undertaken with the aim of establishing a definitive diagnosis prior to intervention.

Nabours et al. classified it as a type 1b extradural meningeal cyst.² ASM is an unusual condition and one article which aims to quantify its prevalence states that there have been 183 cases reported in the postnatal literature from 1837 to 1991 and a further 24 cases from 1990 to 2000.³

The classification of possible causes of ASMs has been described by North et al.⁴ as (1) Congenital (sacral bone defect, proliferation of arachnoid or secondary to connective tissue disorders). (2) Degenerative (resulting from an ischaemic lesion) (3) Traumatic (nerve root avulsion or haemorrhage) or (4) Iatrogenic (usually secondary to orthopaedic surgery).

In contrast to posterior sacral meningocoele (PSM) which classically present in the neonatal age group, anterior sacral meningoceles generally present in adults. There is a strong female predominance with males accounting for only 21% of cases.⁵ Complications include chronic constipation, urinary disturbance, dysmenorrhoea, lower back pain, reduced tone of the anal sphincter and obstruction during childbirth.⁶ Life-threatening emergencies include bacterial meningitis (as was the case with our patient). The radiological investigation of choice to definitively diagnose an anterior sacral meningocele is MRI pelvis.⁷ However, the ‘Scimitar’ sign (a smooth curved unilateral sacral defect) which can often be clearly demonstrated on CT is considered pathognomonic of ASM⁸ thus precluding the need for further imaging.

In fit healthy patients surgery is the mainstay of treatment as myelomengingoceles do not regress spontaneously. It has been suggested that conservative management is associated with up to 30% mortality rate (largely due to bacterial meningitis).⁹ The aim of surgery is to obliterate communication between the meningocele and the spinal subarachnoid space and to decompress the pelvic structures by meningocele excision.^10,11 The standard approach is a posterior sacral laminectomy which allows ligation of the base, to disrupt its connection with the thecal sac. In our patient, the ASM measured 16 cm × 10 cm × 8 cm. Therefore an open approach was offered by the neurosurgical team but subsequently declined. There has been no description in the literature of operative approach specific to management of infected myelomengocoele. It is likely that had our patient consented to operative intervention, he would have undergone incision and drainage of the ischiorectal component, (taking care not to disrupt the myelomeningocoele) followed by interval semi-elective definitive neurosurgical intervention as described above. The most comprehensive account of a stepwise approach to myelomeningocoele¹² describes the following technique. The basic principles are as follows: (1) Coagulation is reduced to minimum using a bipolar diathermy to reduce ischemia, wound breakdown, CSF fistula and meningitis. (2) All nerves fibres are conserved as much as possible. (3) The spinal canal is protected from blood seeping in by gentle packing with saline wet gauze. (4) Skin is dissected off the neural tissues to avoid dermoid and epidermoid implantation cysts. An incision is made between the arachnoid of the neural placode and the zona epithelioza and extended down to the dura. The cuff of skin around the placode is incised and the dura is opened. Intravenous fluid infusion is increased to compensate for cerebrospinal fluid loss. The skin surrounding the placode is excised. The filum terminale is divided to reduce the chance of cord tethering. Closure is in five layers: (1) Reconstruction of the placode to approximate the neural junction. (2) Dural closure. (3) Fascial closure. (4) Subcutaneous tissue. (5) Skin. Large lesions may be closed with z-plasties if skin cannot be easily approximated. Common complications are wound dehiscence, cerebrospinal fluid leakage (as was the concern in our patient), retrograde infection, meningitis and tethered cord syndrome.¹³

The present case is unusual for several reasons. As described, only 21% of ASMs occur in males. While there has been one case previously described in the Japanese literature (female)¹⁴ and another from Spain (female)¹⁵ of an ASM in continuity with a rectothecal fistula, there has never been a description of continuity with an ischiorectal abscess. This communication has clear implications for ascending sepsis from the gastrointestinal tract to the central nervous system. Because of the low incidence and prevalence, a majority of general surgeons would never be faced with management of acute presentation of ASM and so it was very difficult to practice evidence based medicine in establishing the optimum treatment. Similarily, when MRI scans demonstrated communication with the spinal canal, we were reluctant to intervene in a non-neurosurgical centre, without any expertise in operating on meningeal or spinal tissues. While definitive neurosurgical intervention was offered to this patient, the relatively unattractive side effect profile (sepsis, bleeding, and paralysis) caused him to opt for defunctioning ileostomy and a conservative approach. Follow-up MRI demonstrated a marked reduction in the size of the myelomeningocele and resolution of the abscess (following spontaneous drainage and appropriate antibiotic treatment). While ischiorectal abscesses generally present acutely and are emergently incised and drained, we would advocate judicious management and imaging prior to intervention when such abscesses are associated with an unusual constellation of symptoms.

Conflict of interest statement

None.

Funding

None.

Ethical approval

Written informed consent was obtained from the patient for publication of this case report and accompanying images. A copy of the written consent is available for review by the editor in chief of this journal on request.

Author contributions

Aisling Hogan, writing and editing manuscript; Maria Mannion, writing and editing manuscript; Ronan Ryan, provision of images and image interpretation; Waqar Khan, writing and editing; Ronan Waldron, conceptualisation of idea; Kevin Barry, conceptualisation of idea, writing and editing manuscript.