Rectal prolapse is a socially debilitating condition, characterized by rectal protrusion through the anus [1]. Complete rectal prolapse is full-thickness rectal wall prolapse. It is usually associated with a long history of constipation, and with varying degrees of faecal incontinence [2,3]. For this reason, when surgeons select which operational procedure to use, they should consider not only functional results following surgery but also other facts about the patient such as age, sex, comorbidity and the extent of prolapse. In addition, the recurrence, morbidity and mortality rates associated with each procedure need to be considered. Although more than 100 different procedures for the treatment of complete rectal prolapse were described in the 20th century [4,5,6], the procedures fall into two broad categories, according to whether the route of access is abdominal or perineal [6]. Complete rectal prolapse has been studied in various subjects, and some studies have compared abdominal and perineal procedures [1,6]. Abdominal procedures (APs) generally offer lower rates of recurrence but are associated with higher morbidity rates than perineal procedures [1,3,7]. These results might help to select suitable operative procedures for patients with complete rectal prolapse.

In this study we aimed to evaluate and compare the patient factors, operative and functional outcomes of both procedures used to treat patients with complete rectal prolapse.

The medical records of all patients who had primary operation for complete rectal prolapse at a tertiary referral hospital and a university hospital between March 1990 and May 2011 were reviewed retrospectively. The study was approved by the Institutional Review Boards.

Of 127 patients, 19 (9 abdominal and 10 perineal) were excluded because follow-up was less than six months and they could not be contacted by telephone and 4 who underwent Thiersch's operation were also excluded. Follow-up data were obtained from medical records and from standardized telephone interviews in May and June 2012. For patients who had died or whose personal information was lost at the time of the telephone interview, we used the data recorded at the previous follow-up.

The 104 eligible patients were divided into two groups: those who had an AP (n = 64) and those who had a perineal procedure (PP) (n = 40). The operations were performed by four colorectal surgeons (A, B, C, and D). The indication of surgical treatment was the patient's distress related to complete rectal prolapse in the entire patient. The APs included open and laparoscopic rectopexy (modified Ripstein's method) with or without resection. The modified Ripstein's method was performed as follows; posterior fixation of the knitted monofilament polypropylene mesh (Marlex mesh, C.R. Bard, Inc., Murray Hill, NJ, USA) to the sacrum with attachment of the ends of the mesh to the rectum laterally [8], in open as well as laparoscopic surgery. The PPs included Delorme's and Altemeier's procedures.

The operative outcomes analyzed were operation time, length of hospital stay, complications, and recurrence rate. Recorded functional outcomes were constipation, incontinence and manometric results. Constipation was evaluated using the scoring system of Agachan et al. [9], and faecal incontinence was evaluated using the Wexner score [10]. The constipation and incontinence outcomes were categorized as: none, improved, same, or worse. A same or worse outcome was defined as persistent constipation or incontinence compared with the preoperative condition. Of 70 patients at a tertiary referral hospital, 36 patients agreed with the manometric assessment and manometric values were measured at two time points, preoperative and postoperative period. Postoperative measurement was usually checked between 3 and 6 months after operation. The complications were divided into four categories, based on their timing and severity: immediate major or minor complications (within two months of operation), and delayed major or minor complications (two months or more after operation). Major complications were defined as those requiring surgical intervention or hospitalization. Minor complications were defined as those not requiring surgical intervention or hospital admission. Recurrence of rectal prolapse at the follow-up examination or telephone interview was defined as either a recurring full-thickness prolapse, or the recurrence of prolapsed symptoms.

The patient groups were compared using the chi-square test for proportions or Fisher exact test, as appropriate. The pre- and postoperative manometric values were compared using an unpaired Student t-test. Univariate analysis using binary logistic regression was used to assess the risk of delayed major complication. The 5-year recurrence-free period (RFP) was calculated using the Kaplan-Meier method and compared using the log rank test. Statistical significance was defined as P <0.05, and all analyses were performed using the IBM SPSS ver. 19.0 (IBM Co., Armonk, NY, USA).

Several operational procedures for complete rectal prolapse have been performed in our institution over the past two decades. Delorme's procedure was the most commonly performed in our institution throughout the entire study period. By contrast, the type of AP has changed during the last decade because laparoscopy has been increasingly used since 2002: two laparoscopic procedures (22.2% of all APs) were introduced in 2002 and half of the patients who underwent AP in 2010 had laparoscopic procedures. Although several different procedures were performed, the current study compared the AP and PP groups.

Before the comparison of the AP group and the PP group, the current study showed that the relationship between gender and the incidence of rectal prolapse in Korea was very different to that in the West. According to a western review [11], rectal prolapse is more common in women, and more than 80% of patients are female. A Korean study, however, reported similar incidences in men and women [12]. Among the patients who were treated surgically for complete rectal prolapse in the current study, there were similar numbers of females and males (female/male ratio, 52.8%/47.2%), but the proportion of women among the elderly patients aged 70 or older was about 80% (female/male ratio, 80.6%/19.4%). This means the incidence in young male patients is relatively high in Korea. Another Korean study demonstrated that the age of onset for male was lower, the duration of symptoms was longer, and surgery in male was performed younger than in female (mean age, 51 years vs. 64 years) but the severities of incontinence and constipation were lower in male [13]. What makes this difference in incidence among the Korean and the Western is not clear though we suspected the life style which apt to strain in one of the reason. Further study is needed to investigate the incidence of complete rectal prolapse in Korea because the current study was not population-based.

In the current study, the AP group was younger and had a longer prolapsed length of rectum than the PP group. The choice of approach was generally influenced by a clinical preference for a PP because of age and/or pre-existing conditions [7]. Our data suggest that in our institution the age of the patient and the prolapsed length were major considerations in selecting the operation type, as in other studies [6,11,14]. However, there was no significant association between the four complication categories and patient factors (age, sex, reducibility, comorbidity, prolapsed length, duration of disease, main reason for operation, and operation types).

Patients in the PP group had shorter hospital stays, shorter operation times, and higher recurrence rates than patients who had an AP, as shown in previous studies [3,4,15,16,17,18,19]. The PP group was followed up for a shorter time (a median of 4 months less than the AP group), so the recurrence rate at set intervals would be higher for the PP patients. When we analyzed the relationship between the recurrence and the operative approach, the current study showed that recurrences were correlated with the surgeon's early experience in the AP group and the operative method in the PP group. Our observations suggest that surgeons' increased familiarity with APs (both open operations and laparoscopy) might be associated with lower recurrence. Some studies [20,21] suggested that the learning curve for AP might be responsible for recurrences. In the PP group, the recurrence rates are consistent with the rates that have been reported previously: 0%-20% for Altemeier's procedure [22], and 4%-38% for Delorme's procedure [11].

In the current study, AP combined with resection was a risk factor of the delayed major complications; six of the seven patients (85.7%) experienced delayed major complications (rectopexy only vs. resection; odd ratio, 0.19). Resection procedures are reported to significantly reduce constipation in patients who complain of this symptom before surgery [4], but surgeons should be aware of the tendency for delayed major complications.

Persistent constipation was more common than persistent incontinence in the AP group, although this was not statistically significant. In contrast, persistent incontinence tended to be more frequent than persistent constipation in the PP group. This may indicate that the presence of constipation or incontinence before surgery influenced the choice of operative approach. Especially, surgeon would note higher rate of persistent incontinence in the PP group than in the AP group.

In the current study, manometric measurements revealed that the pre- and postoperative measurement of all manometric values except MRP value remained unchanged in both groups and the postoperative MRP values were still below the normal MRP range. Therefore, this study could not suggest whether manometric studies for complete rectal prolapse could predict functional outcomes. A recent study [23] suggested that preoperative manometry could predict the fecal continence rate after proctectomy, because patients with MSP above 60 mmHg had significantly better outcomes. However, it should be noted that there was no long-term physiological follow-up of patients [23]. Further research might be needed to determine a value of physiological study as a predictor of functional outcome.

A study [24], in the early 1990s, insisted that the AP would be performed when PP was failed in the results obtained from younger patients. Recently, laparoscopic ventral rectopexy (LVR) has gained wide acceptance [25] as the surgical treatment of rectal prolapse associated with low complication and recurrence rate [26]. The results of the current study might support LVR partly because the current study showed high incidence of young patients in Korea, relatively higher 5-year RFP in the AP group, and comparable complications. However, the PP would be preserved for the selected patients who should avoid general anesthesia although comparable study [27] was existed.

The present study has a number of limitations that should be considered when interpreting the results. As in all retrospective observational cohort studies, there is potential for both referral and selection bias. The assessment of patients by telephone without physical examination was subjective; although it used detailed questionnaires, it was likely that recurred prolapse was underestimated by the patients. The subgroup analysis of manometric finding was restricted by the small number of patients and the limited implementation of manometry. Intervention groups in the current study were heterogeneous. For this reason, additional comparison between open posterior rectopexy and Delorme's procedure was performed. The current study, however, would provide insight into changes in procedures over time, factors influencing the choice of operative procedure, and factors affecting recurrence.

In the operative management of complete rectal prolapse, there was no significant association between major or minor complications and patient factors. The AP group had similar postoperative complications compared with the PP group except AP combined with resection for a risk of delayed major complications. The overall recurrence in the PP group tended to occur frequently, compared with the AP group. Persistent constipation was more frequent in the AP group, while, persistent incontinence was more prominent in the PP group. Preoperative functional deficit and the early experience of surgeon for AP procedure, accompanied with age, comorbidities, and the length of the prolapsed rectum, would be considered for the selection of surgical procedures.