International Journal of Colorectal Disease https://doi.org/10.1007/s00384-018-2971-4
ORIGINAL ARTICLE
Trends in the treatment of rectal prolapse: a population analysis A. C. Rogers 1 & N. McCawley 1 & A. M. Hanly 2 & J. Deasy 1 & D. A. McNamara 1,3 & J. P. Burke 1,3 Accepted: 29 January 2018 # Springer-Verlag GmbH Germany, part of Springer Nature 2018
Abstract Purpose Rectal prolapse is a common condition, with conflicting opinions on optimal surgical management. Existing literature is predominantly composed of case series, with a dearth of evidence demonstrating current, real-world practice. This study investigated recent national trends in management of rectal prolapse in the Republic of Ireland (ROI). Methods This population analysis used a national database to identify patients admitted in the ROI primarily for the management of rectal prolapse, as defined by the International Classification of Diseases, 10th Revision (ICD-10). Demographics, procedures, comorbidities, and outcomes were obtained for patients admitted from 2005 to 2015 inclusive. Results There were 2648 admissions with a primary diagnosis of rectal prolapse; 39.3% underwent surgical correction. The majority were treated with either a perineal resection (47.2%) or an abdominal rectopexy ± resection (45.1%). The populationadjusted rate of operative intervention increased over the study period, from 25 to 42 per million (p < 0.001), with no change in the mean age of patients over time (p = 0.229). The application of a laparoscopic approach increased over time (p = 0.001). Patients undergoing an abdominal rectopexy were younger than those undergoing a perineal procedure (64.1 ± 17.3 versus 75.2 ± 15.5 years, p < 0.001) despite having a similar Charlson Comorbidity Index (p = 0.097). The mortality rate for elective repair was 0.2%. Conclusions Despite the popularization of ventral mesh rectopexy over the study period, perineal resection Delorme’s procedure remains the most common procedure employed for the correction of rectal prolapse in the ROI, with specific approach determined by age. Keywords Rectal prolapse . Charlson Comorbidity Index . Rectopexy
Introduction Rectal prolapse (or procidentia) generally occurs after the fifth decade, due to pelvic floor weakening, and affects females predominantly [1, 2]. While the incidence is reportedly low, it remains a condition commonly seen in colorectal specialist clinics [1] and often occurs in conjunction with urinary bladder or gynecological organ prolapse [3]. Surgery offers the
* J. P. Burke
[email protected] 1
Department of Colorectal Surgery, Beaumont Hospital, Dublin 9, Ireland
2
Department of Colorectal Surgery, St Vincent’s University Hospital, Elm Park, Dublin 4, Ireland
3
Royal College of Surgeons in Ireland, 123 St Stephens Green, Dublin 2, Ireland
only curative treatment, with many surgical approaches described. The variety of operative approaches is likely because no one repair is suitable for all patients and a robust evidence base is lacking [2]. The two commonest approaches for repair of rectal prolapse are transabdominally and via the perineum. Abdominal surgery may be open or minimally invasive and generally aims to plicate the rectum to the sacrum (rectopexy), thereby limiting its redundant length prolapsing through the anus [4]. The fixation may be primarily sutured or with the assistance of a prosthetic mesh [5]. Rectopexy may be undertaken with or without a coexisting resection [6]. Perineal surgery resects redundant rectal mucosa (Delorme’s procedure) or fullthickness rectal resection (Altemeier’s procedure) from below and can be performed with or without levatorplasty [7]. The advantage of the perineal approach is that it can be undertaken under spinal anesthesia, thus is a good option for those infirm patients unfit for a general anesthetic, and however has a higher risk of recurrence [8].
Int J Colorectal Dis
Ultimately, the choice of surgical approach depends on patient factors (operative risk, patient preference, and expected outcomes) and surgeon preference. The current study aimed to examine the current practices of Irish surgeons treating rectal prolapse using data obtained from a national patient database to determine trends in incidence, most common procedures performed, and the effect of comorbidity on choice of procedure.
reporting is mandatory in all public and most private hospitals in the state. The database uses the International Classification of Diseases, 10th Revision (ICD-10) to categorize each patient admission as a single record, first categorizing the Bprimary diagnosis^ (i.e., the reason for hospital admission), followed by cataloging any pre-existing comorbidities. Inpatient events and procedures are also all collected and logged (Table 1). Quoted figures for mortality refer to death occurring on that inpatient admission.
Materials and methods
Patient cohort
Data retrieval
The HIPE database was interrogated for all inpatient admissions with a primary diagnosis of rectal prolapse (ICD-10 code K623) from 2005 to 2015 inclusive. Pediatric admissions (those under 15 years) were excluded. Patient ages were categorized into 5-year age groups by the HIPE coding system,
The Hospital In-Patient Enquiry (HIPE) system is a digitalized health resource for the collection of demographic and medical data on patient admissions in the Republic of Ireland;
Table 1 ICD-10 codes for procedures relevant to rectal prolapse
Group
ICD-10-AM Code
Procedure
Rectopexy
3211700 3558400 3559501 3211700 3200300
Abdominal rectopexy Repair of pelvic floor prolapse Abdominal pelvic floor repair Abdominal rectopexy Limited excision large intestine with anastomosis
3202600 3202400 3202800 3202500 3206000
Low resection of rectum, coloanal anastomosis High resection of rectum with anastomosis Ultra-low restorative anterior resection Low resection of the rectum (extraperitoneal) Restorative proctectomy
Sphincter repair
9220800 3200300 3212600
Anterior resection level unspecified Limited excision large intestine with anastomosis Sphincteroplasty
Thiersch stitch
3212000
Insertion of anal suture for anorectal prolapse
Perineal mucosectomy
3209900 3213503 3211200 3204700 3210500 3203900 3203000 3037528 3037504 3200601 3200000 3039000 9031400 9031300
Per-anal submucosal excision of rectal tissue Destruction of rectal mucosal prolapse Perineal rectosigmoidectomy Perineal proctectomy Per-anal full thickness excision of anorectal tissue Abdominoperineal proctectomy Rectosigmoidectomy with stoma Temporary colostomy Other colostomy Left hemicolectomy with stoma formation Limited excision large intestine with stoma Laparoscopy Other procedures on rectum Other repair of rectum
Resection rectopexy
(Delorme’s procedure) Perineal rectosigmoidectomy (Altemeier’s procedure) Stoma
Laparoscopy Other
Int J Colorectal Dis
but length of stay was presented as a continuous variable as standard.
Statistical analysis Results were analyzed using SPSS Statistics for Windows (SPSS, Version 22.0. Armonk, NY). Population-adjusted rates were calculated by obtaining Irish population data from the Central Statistics Office, Dublin, Ireland. All listed diagnoses and procedures were categorized for calculation of Charlson Comorbidity Index (CCI) to estimate pre-existing comorbidity (Fig. 1). Individual patient ages were estimated from age groups using Gompertz law [9], and continuous variables were presented as mean ± standard deviation (SD). Comparative analyses of quantitative data were performed using the Student t test or one-way
Fig. 1 Charlson Comorbidity Index score calculator. The score is an additive sum of any comorbidities as shown. AIDS acquired immune deficiency syndrome, HIV human immunodeficiency virus
ANOVA for continuous variables and chi-squared test for categorical variables. Linear regression was used to assess for trends over time. All significance testing used twotailed analysis, and a difference of p < 0.05 was deemed to be significant.
Results Demographic data There were 2647 admissions with a primary diagnosis of rectal prolapse over the study period; 1040 of these (39.3%) underwent 1076 surgical procedures for the correction of prolapse (Fig. 2). Most patients undergoing intervention were female (87.8%), their mean age was 68.1 ± 17.0 years, and
Int J Colorectal Dis Fig. 2 Patient admissions with rectal prolapse in Ireland 2005– 2015. Single asterisk denotes numbers of patients; however, the numbers of procedures will sum to more than the overall number of patients as some patients have multiple procedures
length of stay was 9.1 ± 14.6 days. Of all patients having surgery, 78.4% were admitted for planned elective procedures. There were four inpatient deaths in the procedural group, only one of which resulted from an elective admission (0.4% mortality rate—Table 2).
rectopexy was undertaken with or without resection (20.3 versus 22.2% underwent laparoscopic approach respectively, p = 0.871). Of the remaining patients, 3.9% had a sphincter repair and the remainder received a stoma or had an unspecified rectal repair (Table 3).
Procedures Age Almost half of patients (47.2%) were treated with a perineal approach, 90.5% of these were with a perineal mucosectomy (Delorme’s procedure), and the remainder was with a perineal rectosigmoidectomy (Altemeier’s procedure). A further 469 patients (45.1%) had abdominal rectopexy, of whom 86.4% of these did not have a resection (Table 3). The laparoscopic approach was equally employed independent of whether a
Table 2
Patients undergoing perineal procedures were older than those undergoing rectopexy (mean age for perineal approach 75.2 ± 15.5 years versus rectopexy 64.1 ± 17.3 years, p < 0.001). Of those patients who had a rectopexy, those having resection were older than those who had rectopexy alone (68.3 versus 63.4 years, p = 0.035).
Overview of patients who died following admission for rectal prolapse
Patient
Sex
Age
Year
CCI
Procedure
LOS (days)
Complications
1
F
52
2011
4
Resection rectopexy
45
2
F
72
2005
0
Rectopexy
14
3
F
82
2005
4
Rectopexy
67
4
F
82
2007
1
Delorme’s procedure
35
Hemorrhage requiring re-laparotomy and stoma, wound dehiscence, multiorgan failure with sepsis Postoperative cardiac arrest, ischemic small bowel requiring resection, MRSA septicemia Postoperative respiratory failure with pleural and pericardial effusion and pulmonary edema MRSA septicemia, pneumonia, Alzheimer’s dementia
Int J Colorectal Dis Table 3
Overview of procedures Procedures, n (%) 1076
Mean age (years)
Charlson Comorbidity Index
Length of stay
469 (43.6) 405
64.1 ± 17.2 63.4 ± 17.3
0.3 0.3
9.9 ± 11.5 9.6 ± 11.6
With resection
64 493 (47.4)
68.3 ± 16.5 75.2 ± 15.5
0.3 0.4
11.7 ± 10.6 10.4 ± 17.4
Mucosectomy (Delorme’s procedure)
446
71.5 ± 15.7
0.4
8.3 ± 18.0
Full-thickness resection (Altemeier’s procedure)
47
77.1 ± 11.5
0.7
9.1 ± 8.9
41 (6.4)
75.2 ± 13.9
0.6
10.5 ± 22.0
6 35
75.0 ± 12.5 75.2 ± 14.3
0.2 0.7
7.3 ± 6.2 11.0 ± 23.7
38 (5.9) 35 (5.4)
73.7 ± 14.1 60.4 ± 19.6
0.8 0.1
18.0 ± 15.4 4.7 ± 5.7
Rectopexy Without resection Perineal approach
Sphincter repair Primary repair Thiersch suture Stoma Other
Charlson Comorbidity Index
Length of stay
The mean CCI among all patients undergoing procedures was 0.4. Patients undergoing rectopexy ± resection had a similar CCI to those undergoing perineal prolapse repair (0.3 versus 0.4, p = 0.097) or sphincter repair (0.3 versus 0.6, p = 0.090). Patients undergoing Altemeier’s procedure tended to have a higher CCI than those undergoing Delorme’s (0.7 versus 0.4, p = 0.027). Patients who received a stoma were more likely to have higher comorbidities than all others undergoing procedures (0.8 versus 0.3, p = 0.001).
Patients who required a stoma had the longest length of stay (LOS) at 18.0 days. There were no statistically significant differences in LOS between those undergoing rectopexy ± resection or perineal procedures (p = 0.132).
Fig. 3 Time trends of rates of surgical intervention, age, laparoscopy, and LOS of included patients
Trends over time The population-adjusted rate of operative intervention for rectal prolapse increased over the study period, from 25 to 42 per million (p < 0.001), with no change in the mean age of patients
Int J Colorectal Dis
over time (p = 0.229, Fig. 3). The rates of all approaches (rectopexy ± resection, perineal resections, and sphincter repairs) increased over the study period, but there was no trend towards one or other procedure over time. The rate of laparoscopy also significantly increased over the time period (p = 0.001), while the length of stay overall significantly decreased from 11.7 days in 2005 to 8.1 days in 2015 (p = 0.041).
Discussion and conclusions Rectal prolapse is a common condition but suffers from a dearth of evidence in determining its optimal management. Although there are over 15 randomized controlled trials (RCTs) on the surgical management of this condition, they suffer from a lack of procedural standardization and thus comparability [6, 10–15]. In fact, many of the results offer equivocality, and the consensus is that no one procedure is superior [2, 16, 17]. The current study reports national population data on rectal prolapse surgery obtained during a time period when laparoscopy was exponentially growing in popularity. The PROSPER study (the largest RCT for the surgical correction of rectal prolapse to date) did not comment on the proportion of patients in whom a laparoscopic approach was utilized when undergoing rectopexy [15]. However, some insight was gained when the Association of Coloproctology of Great Britain and Ireland (ACPGBI) surveyed its consultant members in 2014 and demonstrated a reported increasing preference for the abdominal approach to rectal prolapse repair among surgeons in recent years. The assumption was made in their discussion that this was presumably due to the popularization of laparoscopic surgery, although this was not explicitly cataloged in the survey [16]. In the current data, a surge in laparoscopy was described; however, this did not translate to a tendency for Irish surgeons to prefer the abdominal to the perineal approach during the study period. In fact, there was consistently almost an even split between the two approaches year-on-year over the study period. However, as one would expect, the laparoscopic approach was more frequently employed than that of open surgery in the latter years and is associated with improved short-term outcomes [14]. The ACPGBI study also hypothesized that the relative reduction in abdominal resections seen among their members may have coincided with the increase in laparoscopy, and suggested an explanation that laparoscopic resection was more technically challenging than open resection [16]. Interestingly, our results suggest that when undertaking a resection rectopexy, laparoscopic access is as acceptable to Irish surgeons as open access, with no significant difference in the resection rates between the two approaches in this unselected cohort.
Patients undergoing an abdominal rectopexy were younger but had a similar CCI than those undergoing a perineal approach; this suggests surgeon preference for the abdominal approach in younger patients, but not necessarily correlating with comorbidity status. This trend has been demonstrated previously as older patients are seen as a higher operative risk [16]. However, we know that the age of the patient does not correlate with postoperative morbidity, rather the approach [18]. The limitations of the current dataset include an inability to document postoperative complications and recurrence rates, but the mortality rates among both perineal and abdominal approaches are low at 0.2%. Further clarity is needed for those older patients who might undergo rectopexy with better functional outcomes; if the abdominal approach had an acceptable risk profile in the elderly, then it might be safely employed irrespective of patient age. Another finding in this dataset was that those having a stoma fashioned were more likely to be older and have more comorbidities (i.e., have the highest operative risk). Although these patients were also more likely to have a longer length of stay, it is difficult to comment on whether this was due to preexisting comorbidity, procedural-related complications, or time spent for ostomy education. There are inherent limitations associated with the use of epidemiological data to examine conditions such as rectal prolapse. The retrospective nature of the data is susceptible to selection bias, and the quality is determined by the quality of the inputted data. In this study, there are specific challenges—most prominently, we were unable to categorize rectal prolapse by Oxford classification [19] and thus reported on the group as a whole. It was not possible to comment on the technicalities regarding rectopexies, such as ventral versus posterior and mesh use versus suture, as these were not coded separately. ICD procedure coding does not isolate Blaparoscopic rectopexy^ as a procedure code, rather relying on the coder to separately input the code for Blaparoscopy^; this may have led to under-reporting of the laparoscopic approach. The Altemeier group procedural codes could have incorporated other perineal rectosigmoidectomy procedures such as the stapled transanal rectal resection (STARR) procedure or perineal stapled resection, although we anticipate that would encompass a negligible group statistically, since this was only popularized in Ireland in the latter years of the study. Patients in Ireland do not have a unique national identifier number, and this limits the ability of the database to account for multiple admissions or readmissions. Finally, data such as these are not granular enough to determine specific procedure-related complications. While population databases have limitations, this study reports one of the largest datasets of patients undergoing admission for rectal prolapse and reflects a national case-mix over a recent 11year time period. The strength of this data is that it provides
Int J Colorectal Dis
an unselected snapshot of national trends in management, allowing analysis of real-world management strategies outside of controlled trials. This study demonstrates that rectal prolapse surgery is increasingly undertaken in the Irish population, with laparoscopy more often being employed. The Delorme’s procedure remains the most commonly performed procedure. Inpatient mortality rates for the correction of rectal prolapse are low, and the choice of an abdominal approach corresponds with patients being younger. Appropriate patient selection for the variety of operative techniques remains paramount for the surgical management of patients with rectal prolapse. Author contributions All authors meet the criteria for authorship as per ICMJE recommendations.
Compliance with ethical standards
9.
10.
11.
12.
13.
Conflict of interest The authors declare that they have no conflict of interest.
References 1.
Kairaluoma MV, Kellokumpu IH (2005) Epidemiologic aspects of complete rectal prolapse. Scand J Surg 94(3):207–210. https://doi. org/10.1177/145749690509400306 2. Tou S, Brown SR, Nelson RL (2015) Surgery for complete (fullthickness) rectal prolapse in adults. Cochrane Database Syst Rev 11:CD001758. https://doi.org/10.1002/14651858.CD001758.pub3 3. MacLennan AH, Taylor AW, Wilson DH, Wilson D (2000) The prevalence of pelvic floor disorders and their relationship to gender, age, parity and mode of delivery. BJOG 107(12):1460–1470 4. Mercer-Jones MA, D'Hoore A, Dixon AR, Lehur P, Lindsey I, Mellgren A, Stevenson AR (2014) Consensus on ventral rectopexy: report of a panel of experts. Color Dis 16(2):82–88. https://doi.org/ 10.1111/codi.12415 5. Duthie GS, Bartolo DC (1992) Abdominal rectopexy for rectal prolapse: a comparison of techniques. Br J Surg 79(2):107–113 6. Luukkonen P, Mikkonen U, Jarvinen H (1992) Abdominal rectopexy with sigmoidectomy vs. rectopexy alone for rectal prolapse: a prospective, randomized study. Int J Color Dis 7(4):219– 222 7. Elagili F, Gurland B, Liu X, Church J, Ozuner G (2015) Comparing perineal repairs for rectal prolapse: Delorme versus Altemeier. Tech Coloproctol 19(9):521–525. https://doi.org/10.1007/s10151-0151337-y 8. Riansuwan W, Hull TL, Bast J, Hammel JP, Church JM (2010) Comparison of perineal operations with abdominal operations for
14.
15.
16.
17.
18.
19.
full-thickness rectal prolapse. World J Surg 34(5):1116–1122. https://doi.org/10.1007/s00268-010-0429-0 Neilson S, Robinson I (1993) Reinterpreting mortality statistics: some uses of Gompertzian analysis in epidemiological research. J Clin Epidemiol 46(9):1063–1069 Karas JR, Uranues S, Altomare DF, Sokmen S, Krivokapic Z, Hoch J, Bartha I, Bergamaschi R, Rectal Prolapse Recurrence Study G (2011) No rectopexy versus rectopexy following rectal mobilization for full-thickness rectal prolapse: a randomized controlled trial. Dis Colon rectum 54(1):29–34. https://doi.org/10.1007/DCR. 0b013e3181fb3de3 Makela-Kaikkonen J, Rautio T, Paakko E, Biancari F, Ohtonen P, Makela J (2016) Robot-assisted vs laparoscopic ventral rectopexy for external or internal rectal prolapse and enterocele: a randomized controlled trial. Color Dis 18(10):1010–1015. https://doi.org/10. 1111/codi.13309 Novell JR, Osborne MJ, Winslet MC, Lewis AA (1994) Prospective randomized trial of Ivalon sponge versus sutured rectopexy for full-thickness rectal prolapse. Br J Surg 81(6):904– 906 Renzi A, Brillantino A, Di Sarno G, D'Aniello F, Giordano A, Stefanuto A, Aguzzi D, Daffina A, Ceci F, D'Oriano G, Mercuri M, Alderisio A, Perretta L, Carrino F, Sernia G, Greco E, Picchio M, Marino G, Goglia A, Trombetti A, De Pascalis B, Panella S, Bochicchio O, Bandini A, Del Re L, Longo F, Micera O (2011) PPH-01 versus PPH-03 to perform STARR for the treatment of hemorrhoids associated with large internal rectal prolapse: a prospective multicenter randomized trial. Surg Innov 18(3):241–247. https://doi.org/10.1177/1553350611412794 Solomon MJ, Young CJ, Eyers AA, Roberts RA (2002) Randomized clinical trial of laparoscopic versus open abdominal rectopexy for rectal prolapse. Br J Surg 89(1):35–39. https://doi. org/10.1046/j.0007-1323.2001.01957.x Senapati A, Gray RG, Middleton LJ, Harding J, Hills RK, Armitage NC, Buckley L, Northover JM, Group PC (2013) PROSPER: a randomised comparison of surgical treatments for rectal prolapse. Color Dis 15(7):858–868. https://doi.org/10.1111/codi.12177 Gunner CK, Senapati A, Northover JM, Brown SR (2016) Life after PROSPER. What do people do for external rectal prolapse? Color Dis 18(8):811–814. https://doi.org/10.1111/codi.13255 Murphy PB, Wanis K, Schlachta CM, Alkhamesi NA (2017) Systematic review on recent advances in the surgical management of rectal prolapse. Minerva Chir 72(1):71–80. https://doi.org/10. 23736/S0026-4733.16.07205-9 Fleming FJ, Kim MJ, Gunzler D, Messing S, Monson JR, Speranza JR (2012) It’s the procedure not the patient: the operative approach is independently associated with an increased risk of complications after rectal prolapse repair. Color Dis 14(3):362–368. https://doi. org/10.1111/j.1463-1318.2011.02616.x Collinson R, Cunningham C, D'Costa H, Lindsey I (2009) Rectal intussusception and unexplained faecal incontinence: findings of a proctographic study. Color Dis 11(1):77–83. https://doi.org/10. 1111/j.1463-1318.2008.01539.x