Bleeding following TAE can happen in the immediate post-operative period or later. It has been reported in 10–15% of patients following TAE. Nivatvongs et al. [2] reported an incidence of 3% in their series of 72 patients. In a large case series of 100 patients treated with TAE at St Marks hospital, 3 patients (3%) developed hemorrhage [3]. Similarly, of the 117 patients who underwent TAE for rectal villous adenoma at Ferguson clinic, 8.5% developed hemorrhage following the procedure. Early hemorrhages were treated with return to the operating room and control of bleeding with either suture ligation or cautery. Late hemorrhages did not require any operative intervention [4]. With the improved optics of TEM and better instrumentation, it is not surprising that the incidence of hemorrhage is lower as compared to TAE. In a prospective comparison of TAE and TEM, Winde et al. noted a higher rate of post-operative bleeding in the TAE group, 6% versus 2% [5]. In 89 patients who underwent TAE, Moore et al. reported hemorrhage in 1 patient in the TAE group, whereas no significant bleeding was reported in 82 patients undergoing TEM [6]. In contrast, Langer et al. had a higher rate of blood transfusion in the TEM group as compared to TAE group, 8.9% versus 5.3% respectively, though it did not reach statistical significance [7]. In 260 patients undergoing TEM over a 10 year period, Said et al. reported hemorrhage in 4 (1.4%) patients [8]. Of the 590 patients enrolled in a multi-center TEM Italian study, 8 (1.3%) patients had rectal hemorrhage that required blood transfusion and there were 3 (0.5%) patients with post-operative rectal bleeding that required a repeat TEM procedure for suture control of the bleeding. One patient had intra-operative bleeding that required conversion to open surgery for control of hemorrhage [9]. Transanal Minimally Invasive Surgery (TAMIS) was introduced as an alternative to TEM in 2009 [10]. The main purported advantages of TAMIS over TEM includes lower cost, shorter learning curve, better visualization affording a 360-degree visibility as compared to 220 degrees with TEM, wider operative angle and range of motion of standard laparoscopic instruments [11]. In their initial experience in 50 patients who underwent TAMIS for local excision of benign neoplasms and early rectal cancer, Albert et al. reported delayed hemorrhage in only one patient [11]. Transanal Total Mesorectal excision (TATME) is a new evolving technique for low rectal cancers and utilizes a “bottoms up” approach. Surgeons are still in their learning curves and bleeding can occur if the wrong planes are entered. The pre-sacral veins as well as the lateral pelvic walls are potential places of injury and bleeding from these areas can be difficult to control via the transanal route. In the largest published series of 140 TATME procedures, intra-abdominal bleeding was reported in 2 patients (1.4%) and anastomotic bleeding in 3 patients (2.4%) [46]. In 50 cases of TATME reported by Burke et al. [47], there were no cases of bleeding related to the transanal part of the procedure.

Though there is controversy regarding local excision of early rectal cancer as it does not address the regional mesorectal lymph nodes, local excision of early rectal cancer has significant advantages with decreased morbidity and avoidance of a stoma [12]. Transanal excision is limited to tumors of less than 4 cm in diameter, within 6–8 cm of the anal verge [13, 14]. It has the advantage of no significant learning curve or associated equipment cost [6, 13]. However, multiple studies have shown that TAE is associated with higher probability of positive margins as well as fragmentation of specimen. Christoforidis et al. [15] compared TAE versus TEM in 129 patients with pT1 and pT2 rectal cancers. 16% of TAE specimens had a positive margin versus 2% in TEM, and the rate of specimen fragmentation was also higher‚ 6% versus 0%. Moore et al. also compared TEM versus TAE in 171 patients [6]. TEM was more likely to yield clear margins (90% versus 71%) and a non fragmented specimen (94% versus 65%) compared with TAE. With a mean follow up of 39 months, recurrence was less frequent with TEM, 5% versus 27%. In a national cancer database study, TAE was compared with radical resection, the 5-year local recurrence after local excision was significantly higher than standard excision‚ 12.5% versus 6.9%, P = 0.003 for T1 tumors, and 22.1% versus 15.1% for T2 tumors [16]. Using TAE for rectal cancer, Garcia-Aguilar et al. reported a recurrence rate of 18% for T1 tumors and 37% for T2 tumor at 54 months of follow up [17]. With a median follow up of 55 months, Madbouly et al. had a recurrence rate of 28.8% after TAE of T1 rectal cancers [18]. Mellgren et al. reported an estimated 5-year local recurrence rate of 18% for T1 tumors and 47% for T2 tumors after TAE as compared to 0% for T1 tumors and 6% for T2 tumors after radical resection [19]. Similarly, Chorost et al. reported 31% local recurrence rate after local excision of T1 tumors [20]. In a meta-analysis of TEM versus TAE, TEM had a statistically significant lower rate of negative microscopic margins, reduced rate of specimen fragmentation and recurrence compared with transanal excision [21]. In a retrospective study comparing TEM versus anterior resection versus TAE for rectal lesions (adenoma and carcinoma), there was a higher incidence of incomplete resection (R1 resection) in TAE group (37%) as compared to TEM group (19%). The overall recurrence rate was also higher in the TAE group as compared to TEM and anterior resection (26.3% versus 8.9% versus 3.7%) [5]. The advantages gained due to decreased morbidity and avoidance of a stoma with a TAE is offset by to the high rate of incomplete resection and local recurrence rate. TEM/TAMIS on the other the hand may offer a platform for transanal excision of early rectal cancers with acceptable results. Heintz et al. did not find any significant difference in 5 year survival rate between TEM and radical surgical therapy in patients with low risk T1 carcinoma [22]. A recent meta-analysis of local treatment for T1N0M0 rectal cancer showed that TEM subgroup did not have significantly lower overall survival than radical resection, whereas TAE was associated with significantly lower 5-year overall survival. Additionally, TEM was associated with lower post-operative complications and need for permanent stoma as compared to radical resection [23]. Transanal minimally invasive surgery (TAMIS) was introduced in 2009 and has emerged as a viable alternative to both TEM and TAE [24]. At present there are no studies comparing TAMIS versus standard transanal excision, TEM or radical anterior resection. Systematic review of TAMIS procedures (n = 390 reported a positive margin rate of 4.36% and tumor fragmentation rate of 4.1% [25]. Trials comparing TAMIS versus standard resection are awaited.

Urinary retention is one of the most common complication after anorectal surgery. The incidence varies between 1 and 50% [26–28]. The exact etiology of urinary retention following anorectal procedures is not completely understood. Dysfunction of detrusor muscle or the trigone in response to pain or distention of the anal canal or perineum has been postulated as a cause of urinary retention [29]. Zaheer et al. reported that urinary retention developed in 16% of patients following surgery for benign anorectal conditions [27]. The incidence varied from 34% following hemorrhoidectomy to 4% after lateral internal sphincterotomy and 2% after fistulotomy [29]. The reported rate of urinary retention following TEM, TAMIS, and TATME varies from 5 to 10% [30–32], 0 to 6% [11, 33–35] and 2 to 4%, respectively [46, 47].

Pelvic sepsis following anorectal surgery is fortunately rare, with majority of cases being reported following treatment for hemorrhoids. There are a few case reports of pelvic sepsis following injection sclerotherapy [36–38]. Severe sepsis is more common after rubber band ligation as compared to injection sclerotherapy. A systematic review of life threatening sepsis following treatment for hemorrhoids reported 38 patients with severe sepsis. 17 patients had undergone rubber band ligation of which 6 patients died, 3 patients had undergone excisional hemorrhoidectomy of which 1 patient died and finally, 7 patients with stapled hemorrhoidopexy of which 1 patient died [39]. Kam et al. reported their experience with 7302 stapled hemorrhoidectomy operations in a single center in 2010. They reported 4 cases of perianal sepsis with no deaths [40]. Butterworth et al. treated 118 patients over a period of 4 years with stapled hemorrhoidopexy and reported 1 patient with severe sepsis who eventually died [41]. There are no reliable reports of pelvic sepsis following various methods of TAE.

Anal strictures and stenosis are most commonly seen after hemorrhoidectomy, with an incidence ranging from 1.5 to 3.8% [42]. Multiple systematic review of trials comparing conventional hemorrhoidectomy to stapled hemorrhoidectomy have not shown any statistical difference in the incidence of early and late anal stricture between the two methods [43–45]. None of the larger series of TAE reported anorectal stenosis/stricture. This could be related to the excision and closure of defects in the rectum rather than the anal canal.

Although uncommon urethral injury has been reported following TATME. Rouanet et al. [48] reported 2 urethral injuries in 25 procedures, whereas Burke et al. [47] had one urethral injury in 50 patients. In the largest series of TATME, there were no reported urethral injuries [46].