Rubber band ligation (RBL) is a very commonly undertaken non-operative procedure for the management of internal haemorrhoids first introduced in the 1950’s [2,3]. The procedure is effective, generally well tolerated and mostly proceeds in the absence of any major complications [1,3]. Minor complications of the procedure include; mild bleeding, post-procedural pain, band slippage, anal fissure formation and chronic longitudinal ulceration [2]. These are seen far more commonly than major complications, such as; massive bleeding, thrombosis of the banded haemorrhoids, pelvic sepsis and secondary re-bleed [2]. Rectal ulceration following RBL is considered an extremely rare complication [3]. The exact incidence of rectal ulceration remains unknown [3].

Secondary re-bleeding is rare and in the setting of rectal ulceration is extremely uncommon [3]. The use of anti-platelet medication such as aspirin is associated with greater risk of massive secondary re-bleeding given its inherent ability to impede haemostasis via inhibiting platelet production of thromboxane A2. The irreversible nature of aspirins affect on haemostasis renders the patient to have impaired haemostasis for around 5-7 days following commencement of the medication.

The commencement of anti-platelet medication following cardiac surgery is required in order to protect against thrombogenic phenomena occurring from the graft site. The following describes a case where a young male patient underwent an elective Ross Procedure (Pulmonary Autograph Procedure) and Aortic Root Replacement for the management of symptomatic stenotic aortic disease (secondary to congenitally bicuspid aortic valve) and aortic root replacement for aneurysmal aortic, who ultimately required emergency colorectal surgical management for massive PR bleeding in the setting of rectal ulceration following commencement of anti-platelet therapy post cardiac surgery and RBL for symptomatic bleeding internal haemorrhoids two weeks prior.

Case Presentation

A 45-year-old self-employed caucasian male was electively admitted for the operative management of symptomatic stenotic aortic valve disease and aneurysmal aortic root, undergoing a Ross Procedure and Aortic Root Replacement. His past medical history included Hypertension, Obstructive Sleep Apnoea, Subarachnoid haemorrhage, Gout, Haemorrhoids and Atrial fibrillation (AF). His regular medications included Apixaban, Frusemide, Metoprolol, Rosuvastatin, Amlodipine, Valsartan and Allopurinol. He was an ex-smoker (ceasing three weeks prior to admission), a previous regular consumer of alcohol (ceasing six weeks prior to admission), and an occasional cannabis user. A strong family history of cardiovascular disease was present.

Haemorrhoidal bleeding, complicated bv Apixaban use for AF, was a known issue for our patient prior to booking and undergoing cardiac surgery. For management of the PR bleeding our patient underwent Examination Under Anaesthesia (EUA) + Flexible sigmoidoscopy + RBL + Polypectomy two weeks prior to his planned cardiac surgery. Grade III haemorrhoids were found and four bands were used. A sessile polyp was removed with cold snare from the mid descending colon. Our patients Apixaban were withheld for an appropriate period of time prior to the procedure.

The withholding of the Apixaban continued until planned cardiac surgery which occurred two weeks following the RBL and Polypectomy. The procedure was uncomplicated and as is routine post cardiac surgery in our centre the patient was admitted to the Intensive Care Unit (ICU) for monitoring where he was extubated successfully. He was commenced on aspirin and chemical venous thromboembolic (VTE) prophylaxis on post-operative day one and discharged to the Cardiac High Dependency Unit (HDU) the same day with only minor vasopressor requirements. While in the HDU he was noted to have increasing vasopressor requirements and he was re-admitted to the ICU for further monitoring with the provisional diagnosis of vasoplegia. He was able to successfully wean blood pressure supports in the ICU and was again planned for step down to the HDU.

Overnight on post-operative day three our patient had a massive PR bleed with loss of approximately 750-1,000 ml of frank blood associated with haemodynamic instability. Resuscitation with blood products allowed for return of haemodynamic stability and a decision was made to attend CT angiography. At CT there was found to be active bleeding from the mid-rectum. The patient then attended emergency theatre for EUA + Oversewing of rectal ulcer at the site of RBL. At operation an ulcer at the 9 o’clock position was found above the dentate line with a false aneurysm at the base. Haemostasis of the bleeding was surgically achieved.

The patient returned to the ICU for monitoring post. A further PR bleed with an estimated loss of a further 500-750 ml was noted again resulting in haemodynamic instability. A mix of fresh and altered blood was noted on examination. He was again fluid responsive and taken to theatre for further EUA. The second procedure noted further ooze at a site near the previously noted rectal ulceration. This was managed with the use of a LigaSure device. The remainder of old blood present in the colon was retrieved in order to aid in further decision making in the setting of further PR loss. Aspirin and chemical venous thromboembolism (VTE) prophylaxis was cautiously recommenced following intervention.

Following these events in the ICU our patient was successfully returned to the cardiac HDU and then further stepped down from there. Cardiac issues in the post-operative period included rapid atrial fibrillation which was managed with an Amiodarone infusion and Beta-blocker. A decision was made to not recommence apixaban in the setting of massive PR bleed and performing ligation of the left atrial appendage. Our patient will be followed up by the cardiology department as an outpatient to consider the need for ongoing anticoagulation for AF.

Discussion

Management of symptomatic internal haemorrhoids with RBL is common, effective and is considered to have a low complication rate. Re-bleeding post RBL can occur, usually presenting around 10-14 days following the procedure [4]. Heightened risk of re-bleeding is seen in patients taking anti-platelet medication, with multiple reported cases of massive life-threatening bleeding occurring in this subset of patients [3,5].

One study of 805 patients undergoing RBL found that bleeding was a complication seen 2.8% of the time, with a greater rate of bleeding occurring in those using anti-platelet and / or anti-coagulant medications [6]. In this study, 7.5% of the 40 patients taking aspirin bled post-operatively. It is noted that when bleeding does occur, it generally occurs late after 10 days [7].

A retrospective study that included 364 patients looked at patients on anti-platelet therapy undergoing RBL and the effect of holding anti-platelet therapy 7-10 days following the procedure [8]. Included in the study were 283 patients taking aspirin. A total of 23 bleeding complications were recorded. Of this number, 16 of the patients with bleeding complications were taking aspirin [NR]. Of these bleeding complications in patients taking aspirin 13 were considered insignificant bleeds while three were deemed significant. Overall, the study showed that holding the anti-platelet therapy post-procedure appeared to reduce the risk of bleeding in this subset of patients when compared to patients not taking anti-platelet therapy [8].

Our patient ceased had his anticoagulation for known atrial fibrillation ceased in a appropriate manner prior to the RBL. However, he was heparinised for cardio-pulmonary bypass and commenced on aspirin and VTE prophylaxis as routine care following cardiac surgery occurring14 days post the RBL. He then proceeded to have massive PR bleeding 3 days post his cardiac surgery, timing that correlates with aspirin reaching its maximal therapeutic level, which subsequently required surgical intervention.

With this knowledge regarding the effects of anti-platelet therapy increasing the risk of re-bleeding following RBL, it is not entirely unreasonable to consider that the events that ensued in the post-operative period may have been avoided if the cardiac procedure had been delayed by a further 1-2 weeks following the RBL. It is reasonable to think that a delay like this would have provided our patient with a greater ability to recover from the RBL prior to the commencement of anti-platelet medication. Unfortunately, it is unclear from the correspondence between the surgical treating teams as to whether the relationship and timing of these procedures had been discussed.

Although these two events are likely an uncommon temporal occurrence, it is not unreasonable to imagine that the timing of such procedures (or other procedures that require anti-platelet therapy eg vascular stents or grafts) may again occur in the future. Therefore, it is important for us to review such a case and take from it the importance of considering how our interventions may affect our patients previous or future interventions.