Radiotherapy and Oncology 108 (2013) 55–60

Contents lists available at SciVerse ScienceDirect

Radiotherapy and Oncology

journal homepage: www.thegreenjournal .com

Anal cancer

Faecal incontinence after chemoradiotherapy in anal cancer survivors:Long-term results of a national cohort

0167-8140/$ - see front matter � 2013 Elsevier Ireland Ltd. All rights reserved.http://dx.doi.org/10.1016/j.radonc.2013.05.037

⇑ Corresponding author. Address: Department of Oncology, University Hospital ofNorth Norway, 9038 Tromsø, Norway.

E-mail address: anne.gry.bentzen@unn.no (A.G. Bentzen).

Anne Gry Bentzen a,b,⇑, Marianne G. Guren c, Barthold Vonen b,d, Eva H. Wanderås c, Gunilla Frykholm e,f,Tom Wilsgaard g, Olav Dahl h,i, Lise Balteskard a,j

a Department of Oncology, University Hospital of North Norway, Tromsø; b Department of Clinical Medicine, University of Tromsø; c Department of Oncology, Oslo University Hospital;d Nordland Hospital, Bodø; e Department of Oncology, St. Olav’s University Hospital, Trondheim; f The Norwegian Radiation Protection Authority (NRPA), Østerås; g Department ofCommunity Medicine, University of Tromsø; h Section of Oncology, Institute of Medicine, University of Bergen; i Department of Oncology, Haukeland University Hospital, Bergen; andj Centre for Clinical Documentation and Evaluation, Northern Norway Regional Health Authority, Tromsø, Norway

a r t i c l e i n f o

Article history:Received 8 November 2012Received in revised form 7 May 2013Accepted 30 May 2013Available online 25 July 2013

Keywords:Anus neoplasmChemoradiotherapyFaecal incontinenceRadiotherapy

a b s t r a c t

Purpose: To examine the prevalence and severity of faecal incontinence amongst anal cancer survivorsafter chemoradiotherapy.Material and methods: Anal cancer survivors from a complete, unselected, national cohort, minimum 2-years follow-up, were invited to a cross-sectional study. The St. Mark’s incontinence score was used toevaluate occurrence and degree of faecal incontinence the last four weeks. The results were comparedto age- and sex-matched volunteers from the general population.Results: Of 199 invited survivors and 1211volunteers, 66% and 21%, respectively, signed informed con-sent. The survivors had significantly higher St. Mark’s score than the volunteers (mean 9.7 vs. 1.1,p < 0.001). Incontinence of stool of any degree was reported by 43% vs. 5% (OR 4.0, CI 2.73–6.01), andurgency was reported by 64% vs. 6% (OR 6.6, CI 4.38–9.90) of the survivors and volunteers, respectively.Only 29% of those with leakage of liquid stool used constipating drugs. Survivors of locally advancedtumours had a higher incontinence score (p < 0.01).Conclusions: Moderate to severe faecal incontinence is common amongst anal cancer survivors. Post-treatment follow-up should include the evaluation of continence, and incontinent survivors should beoffered better symptom management and multidisciplinary approach if simple measures are insufficient.

� 2013 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 108 (2013) 55–60

Squamous cell carcinoma of the anal region is a rare malignancywith an incidence of approximately 1 in 100,000 per year [1]. InNorway, 50–60 patients are diagnosed per year. In a recent studyof Norwegian anal cancer patients treated with chemoradiotherapy(CRT), the 5-year cancer-specific survival was 75% [2]. Although thetreatment is effective, a substantial proportion of the survivors re-ported impaired long-term health-related quality of life (HRQOL)with high symptom scores from the pelvis, such as diarrhoea andsexual dysfunction [3]. High symptom scores for faecal inconti-nence suggested that this was a major patient-reported problem.

Faecal incontinence is a well-known late effect after pelvicradiotherapy, but prevalence varies widely, with rates rangingfrom 3% to 53% [4]. Recent published data from smaller studiesof long-term HRQOL of survivors after anal cancer have reporteda high frequency of difficulty in bowel control that might impacton daily life [5–7]. Larger studies are available for survivors afterradiotherapy for other pelvic malignancies [8–10], indicating that

faecal incontinence after pelvic radiotherapy represents a frequentproblem. The prevalence of faecal incontinence depends on the cri-teria used and the heterogeneity of the patient cohort. The use ofstandardised methods enables cross-study comparisons. In thisstudy, the St. Mark’s score was used to evaluate faecal incontinence[11].

The aim of this study was to examine the prevalence and sever-ity of faecal incontinence in the long-term follow-up of anal cancersurvivors after curative CRT.

Materials and methods

Anal cancer survivors

A cohort of all patients diagnosed in Norway between July 2000and June 2007, with squamous cell carcinoma of the anal regionand treated with curatively intended CRT, was included in a na-tional study [2]. Curative treatment of anal cancer was centralisedto five centres. All centres participated in the study. Patient, tu-mour and treatment characteristics were registered in a nationaldatabase. Survivors without any sign of disease recurrence, with

328 pa�ents treated with cura�vely intended (chemo) radiotherapy

115 dead

213 anal cancer survivors

2 survivors with current anal cancer

207 survivors without any sign of anal cancer

4 survivors under inves�ga�on for recurrence

8 survivors not offered par�cipa�on

199 survivors offered par�cipa�on

1 language incompa�bility

7 severe comorbidity

51 non-responders

17 nega�ve response (5 of these par�cipated only in the HRQOL study)

131 signed informed consent

107 anal cancer survivors par�cipated in the study evalua�ng faecal incon�nence

24 stoma

Fig. 1. Inclusion of survivors.

56 Faecal incontinence in anal cancer survivors

a minimum follow-up of 2 years after diagnosis, n = 199, were con-tacted by mail in the period August 2009 to March 2010, and in-vited to participate in a cross-sectional survey. Non-responderswere reminded twice. Inclusion of survivors is shown in Fig. 1.

Volunteers

The survivors’ scores were compared to scores from a group ofvolunteers from the general population, age- and sex-matched atthe time of the follow-up study. These were randomly drawn fromthe National Population Register and invited by mail to participate.The invitation included the information that any history of cancerin the pelvis or abdomen would exclude participation. To obtain atwofold number of volunteers compared to the number of cases ineach sex- and 10-year age group, repeated random draws weredone.

Treatment of anal cancer

According to national guidelines during the study period, therecommended radiation dose was 54–60 Gy and either one courseof Mitomycin-C in combination with 5-Fluorouracil (5-FU) or threecourses of Cisplatin in combination with 5-FU according to tumour

stage. CT-based 3D treatment planning was used in some and grad-ually introduced in all centres in Norway during 2000–2002. Radi-ation was mostly delivered with a 2–4-field technique and 6–18-MV photon beams. Detailed information about the treatment ispresented in a previous paper [2].

Measurements

This cross-sectional survey consisted of a structured telephoneinterview, performed by trained health personnel during the peri-od September 2009 to August 2010. There was a fixed set-up ofpre-defined questions, amongst them questions regarding socio-demographics, comorbidity (diabetes mellitus, chronic inflamma-tory bowel disease, and cardiovascular disease), smoking habits,and the St. Mark’s score of faecal incontinence. Time for the inter-view was agreed upon in advance. The interview lasted approxi-mately 15 min. In addition, the participants completed a HRQOLquestionnaire, the European Organization of Research and Treat-ment of Cancer (EORTC) core questionnaire, QLQ-C30 version 3.0[12], received by mail. The full results of HRQOL are presented ina recent published paper [3]. The age- and sex-matched group ofvolunteers went through the same interview and completed thesame questionnaires as the survivors.

Table 1Patient, tumour, and treatment characteristics of responders and non-responders/excluded anal cancer survivors.

Survivors,responding,n = 107

Survivors, non-respondingor excluded, n = 92

p-Value

Sex 0.154Women 87 (81%) 67 (73%)Men 20 (19%) 25 (27%)

Median age, years 60 (39–89) 70 (40–92) <0.001Medical comorbidity 29 (27%) 47 (51%) 0.001

T-stage 0.057T1 23 (21%) 11 (12%)*

T2 48 (45%) 37 (40%)T3 22 (21%) 21 (23%)T4 14 (13%) 23 (25%)

N-stage 0.369N0 78 (73%) 62 (67%)*

N1-3 29 (27%) 30 (33%)

A.G. Bentzen et al. / Radiotherapy and Oncology 108 (2013) 55–60 57

St. Mark’s score

The St. Mark’s score is a validated instrument to score the fre-quency and degree of faecal incontinence during the last fourweeks [11]. This incontinence score has been found to reflect pa-tients’ perceptions quite well and to be reliable regardless of thetype of incontinence, age or sex [13]. It consists of seven questionsexploring the frequency of involuntary leakage of gas, liquid stool,solid stool, and alteration in lifestyle; need to wear a pad or plug;use of constipating drugs; and the ability to defer defecation for15 min as an indication of urge. The three items of type and fre-quency of incontinence and the item of alteration of lifestyle arescored on a 4-point scale: never = 0, rarely (1 episode) = 1, some-times (>1 episode) = 2, weekly (P1 episodes a week) = 3, and dai-ly = 4. The two items regarding use of pad/plug or constipatingdrugs are binary: no = 0 and yes = 2. The last item regarding ur-gency is binary: no = 0 and yes = 4. The incontinence score rangesfrom 0 (completely continent) to 24 (completely incontinent).

TNM group 0.122T1-2N0M0 60 (56%) 42 (46%)*

T3-4N0/T1-4N1-3M0 47 (44%) 50 (54%)Radiotherapy, median

dose, Gy54 (44–64) 54 (38–66) 0.697

Chemotherapy 101 (94%) 80 (87%) 0.105

TNM clinical classification: T – primary tumour, N – regional lymph node, M –distant metastasis.* 1 Excluded survivor with unknown TN-classification is included in this group.

Table 2Treatment characteristics according to tumour stage.

T1-2N0M0n = 60

T3-4/N0 or T1-4/N+M0 n = 47

Radiotherapy (RT)Median dose in Gy, range (mean) 54, 44–60 (53) 58, 50–64 (57)Radiotherapy field border

Promontory 10 (17%) 37 (79%)Lower border of sacroiliac joint 50 (83%) 10 (21%)

Irradiation of the groins 43 (72%) 41 (87%)Median treatment time of RT, days

(range)36 (25–59) 38 (32–58)

Chemotherapy5-Fluorouracil 55 (92%) 46 (98%)Mitomycin-C 53 (88%) 13 (28%)

Statistics

Data were analysed using SPSS version 18.0 (SPSS Inc., Chicago,IL). Differences between responders and non-responders were ana-lysed using Chi-square tests for categorical variables, and Studentt-tests for continuous variables. The scores from the HRQOL werestandardised by linear transformation into a score ranging from 0to 100 according to the EORTC scoring manual [14]. A high scoreon the global quality of life (QOL)/functional scales represents ahigh QOL or a high level of functioning, and a high score for thesymptom items represents a high level of symptoms. Due to age-and sex-matched volunteers, conditional logistic regression wasused to compare scores and symptoms between survivors and vol-unteers. In analyses of survivors, Mann Whitney U tests were usedto compare St. Mark’s score between subgroups. The faecal incon-tinent factors were dichotomised based on what we considered tobe clinical relevant. Dichotomised faecal incontinent factors wereused as dependent variables in logistic regression models to assessthe associations with sex, age and tumour stage in subgroups ofsurvivors. Both crude and multivariable models (with all threeindependent variables included) were assessed. To assess correla-tion between the mean St. Mark’s score and mean scores of HRQOL,the Spearman’s Rank Order Correlation (rho) was used.

Cisplatin 2 (3%) 40 (85%)

TNM clinical classification: T – primary tumour, N – regional lymph node, M –distant metastasis.

Ethical considerations

All data received were treated in strict confidence. The studywas approved by the Regional Committee for Medical ResearchEthics of Northern Norway.

Results

Participants

Informed consent was signed by 66% of the survivors. Stoma ex-cluded 24 survivors from participation, leaving 107 survivors eligi-ble for the final analysis of faecal incontinence (Fig. 1). Patient,tumour and treatment characteristics of responders and non-responders are described in Table 1. The responders were in gen-eral younger than the non-responders and had less comorbidity.Detailed treatment characteristics of the 107 anal cancer survivorsaccording to stage of disease are shown in Table 2. The median fol-low-up after diagnosis of the 107 anal cancer survivors was66 months (range 25–111).

Of 1211 invited volunteers, 259 (21%) accepted to participate.The prevalence of self-reported diabetes mellitus (4%), chronicinflammatory bowel disease (<1%), and cardiovascular disease(3–7%) did not differ significantly between survivors and volun-

teers. Smoking was more common amongst survivors than volun-teers, 32% vs. 13%.

Faecal incontinence

The mean St. Mark’s score was significantly higher in anal can-cer survivors than volunteers, with a mean score of 9.7 vs. 1.1,respectively, p < 0.001. Anal cancer survivors had scores rangingfrom 0 to 20 (Fig. 2).

The detailed results of the St. Marks score are shown in Table 3.Incontinence for gas (weekly or daily) was significantly more com-mon amongst survivors than volunteers, 55% vs. 12%, (OR 3.9, 95%CI, 2.61–5.73). Incontinence for liquid stool (rarely–daily) occurredin 41% vs. 5% (OR 3.9, 95% CI, 2.61–5.77) and incontinence for solidstool (rarely–daily) occurred in 15% of survivors, but none of thevolunteers, respectively. Incontinence for stool, either liquid or so-lid (rarely–daily), occurred in 43% of the survivors and 5% of thevolunteers (OR 4.0, CI 2.73–6.01). Faecal incontinence resulted inalteration in lifestyle (weekly–daily) in 54% of the survivors com-pared to 3% of the volunteers (OR 5.6, 95% CI, 3.80–8.27). Urgency

Fig. 2. Distribution of St. Mark’s score of faecal incontinence in anal cancersurvivors and volunteers.

58 Faecal incontinence in anal cancer survivors

was reported in 64% of the survivors compared to 6% of the volun-teers (OR 6.6, 95% CI, 4.38–9.90).

Items regarding diarrhoea, role- and social functioning, and glo-bal QOL during the last week were investigated in the self-reportedquestionnaire, EORTC QLQ-C30. Diarrhoea (quite a bit or verymuch) was reported by 21% vs. 6%, and the mean score of diarrhoeawas 29 in survivors compared to 13 in volunteers (p < 0.001). Analcancer survivors had significantly lower role- and social function-ing, mean scores 75 and 76, compared to 92 and 92 in volunteers(p < 0.001). Mean score of global QOL was significantly loweramongst anal cancer survivors, 72, compared to 83 amongst volun-teers (p < 0.001).

Table 3Percentage distribution of St. Mark’s score for survivors and volunteers.

Never (%)

Incontinence for gas Survivors 31Volunteers 73

Incontinence for liquid stool Survivors 59Volunteers 95

Incontinence for solid stool Survivors 85Volunteers 100

Alteration in lifestyle Survivors 20Volunteers 92

Need to wear sanitary pad or plug SurvivorVoluntee

Taking constipating medicines SurvivorVoluntee

Lack of ability to defer defecation for 15 min SurvivorVoluntee

Never: no episodes the past four weeks. Rarely: 1 episode the past four weeks. Sometimesa day. Daily: P1 episodes a day.

Impact of patient and treatment-related factors on faecal incontinence

Weekly or daily unintended leakage of gas was more commonamongst females than males, p = 0.050 in crude analysis andp < 0.05 in analysis adjusted for age and tumour stage (Table 4).Apart from leakage of gas, there were no significant sex differencesregarding incontinence for stool, urgency, alteration of lifestyle,use of constipating drugs or use of pads/plugs. The mean St. Mark’sscore did not differ significantly between the sexes. With increas-ing age, survivors reported greater use of pads/plugs, p < 0.05 inboth crude and adjusted analyses. No other faecal incontinencefactors were significantly associated with age. Surviving locally ad-vanced tumours was associated with significantly more leakage ofstool, more frequent lifestyle alterations, and more use of pads/plug than surviving smaller tumours, p < 0.05 both in crude and ad-justed analyses. The mean St. Mark’s score was significantly higherfor survivors after locally advanced compared to smaller tumours:11.6 vs. 8.3 (p < 0.05). There was no significant difference in prev-alence of faecal incontinence between patients who had receivedirradiation to the groins compared to those without groin irradia-tion. Smoking habits did not influence the occurrence of faecalincontinence.

A moderate correlation was found between the mean St. Mark’sscore and role function (Spearman’s rho�0.301 p < 0.01) and socialfunction (Spearman’s rho �0.254 p < 0.05), but not between themean St. Mark’s score and global QOL.

Management of faecal incontinence

Constipating drugs were used by 18 survivors (17%). Leakage ofliquid stool was reported by 44 survivors (41%) and constipatingdrugs were used by 11 (25%) of these. Use of sanitary pad/plugwas reported by 29 (27%) of the survivors, amongst these 25 re-ported unintended leakage of liquid stool.

Discussion

Survivors of anal cancer have high scores of faecal incontinencewith a mean score of 9.7 on a scale from 0 to 24, indicating that thisis a major problem of anal cancer survivorship. Any degree ofincontinence for stool was reported by 43% of the survivors. Ur-gency was very common and reported by 64%.

Reports on prevalence of faecal incontinence in the general pop-ulation are heterogeneous, and the estimates depend on defini-tions and data collection methods applied [15]. Population-based

Rarely (%) Sometimes (%) Weekly (%) Daily (%)

5 9 19 368 7 8 48 14 13 63 2 0 04 6 3 20 0 0 0

10 16 6 484 1 2 1

No (%) Yes (%)

s 73 27rs 99 1

s 83 17rs 99 1

s 36 64rs 94 6

: >1 episode the past four weeks but <1 a week. Weekly: P1 episodes a week but <1

Table 4Multivariable adjusted odds ratio for components of St. Mark’s score in survivors.

Weekly or dailyincontinence for gas

Incontinence for liquidand/or solid stool*

Weekly or dailyalteration of lifestyle

Need to use pads/plug Constipating drugs Urgency

OR p-Value

CI OR p-Value

CI OR p-Value

CI OR p-Value

CI OR p-Value

CI OR p-Value

CI

Male: female 0.33 0.037 0.12–0.94

0.74 0.576 0.25–2.16

0.74 0.566 0.27–2.06

0.52 0.355 0.13–2.08

2.43 0.159 0.71–8.38

0.95 0.927 0.34–2.68

Age (per10 years)

0.77 0.196 0.52–1.14

0.96 0.841 0.64–1.44

1.20 0.371 0.81–1.78

1.76 0.016 1.11–2.80

1.37 0.226 0.82–2.28

1.45 0.083 0.95–2.22

T3-4N0/N+:T1-2N0

1.33 0.482 0.60–2.95

4.03 0.001 1.78–9.10

2.73 0.015 1.22–6.09

3.47 0.009 1.36–8.84

2.43 0.101 0.84–7.04

1.54 0.307 0.67–3.52

* Rarely, sometimes, weekly or daily.

A.G. Bentzen et al. / Radiotherapy and Oncology 108 (2013) 55–60 59

surveys have reported a prevalence of unintended leakage of stoolvarying from 1% to 6% in the normal non-institutionalised popula-tion [16–18]. Increased age is a risk factor, while female gender as arisk factor is controversial [16,17,19]. Diarrhoea and urge werestrongly associated to faecal incontinence, but not parity, in a largepopulation study of Norwegian women [18]. Tobacco smoking wasfound to increase the risk of long-term bowel- and anal symptomsafter radiotherapy in a study of prostate cancer survivors [20]. Lackof detailed obstetric and smoking history in this study, limited theopportunity to evaluate the importance of these factors in relationto faecal incontinence.

Faecal incontinence has been reported after pelvic radiotherapyfor other cancers. In a cross-sectional study of preoperative radio-therapy in locally advanced rectal cancer, 67% reported inconti-nence of liquid stool and 44% urge [10]. A large study of self-reported symptoms of faecal incontinence amongst gynaecologicalcancer survivors reported defecation urgency at least one a week in29%, and leakage of loose stool while awake in 33% [9].

Rectal urgency and uncontrolled leakage of stool, at least once aweek, was reported by 35% and 15–18%, respectively, in a study oflong-term follow up after curative radiotherapy for prostate cancer[8].

In the current study, the mean St. Mark’s score amongst the sur-vivors was 9.7 vs. 1.1 in the group of volunteers. Interpretation ofthe score is challenging since there exists no established cut-off va-lue that distinguishes normal variation from pathology or for clas-sifying faecal incontinence into moderate and severe. The clinicalimplication of an abnormal score may vary individually. Still,regardless of the total score, leakage of stool is considered to bea pathological condition, and urge represents major problems indaily life. The high mean score in the anal cancer survivors in thecurrent cohort clearly demonstrates that many of them suffer frommoderate to severe faecal incontinence.

The survivors had significantly lower global quality of life scorescompared to volunteers. Due to the nature of the symptoms, anassociation between faecal incontinence and HRQOL could be ex-pected. However, HRQOL is a multidimensional concept and theimpact of faecal incontinence is unclear [7,21]. A moderate nega-tive correlation was found between mean St. Mark’s score andmean scores of role and social function. This might indicate im-paired role- and social function in survivors with severe faecalincontinence. However, causality should be concluded with cau-tion as several factors contribute to the HRQOL scores. EORTCQLQ-C30 is not designed to measure HRQOL in patients with faecalincontinence and a condition-specific questionnaire would proba-bly be more appropriate.

Continence is a complex functional process involving severalfactors including anatomical aspects [4]. The radiotherapy fieldsinclude the rectal ampulla and sacral nerves as well as the pelvicfloor with the anal sphincter. Impaired compliance, motility or sen-

sory- and motor-function may result in dysfunction of the rectalreservoir. The entire anal canal and sphincter is recommended tobe included in the clinical target volume in radiotherapy [22]. Radi-ation induced normal tissue injury could be composed of both tis-sue atrophy due to parenchymal cell damage or loss, andreplacement fibrosis [23]. In the present study, surviving locallyadvanced tumours were associated with more faecal incontinencethan surviving smaller tumours. Larger tumours are usually moreinvasive, potentially causing more damage to the sphincter andanal canal. Locally advanced tumours were also treated moreextensively, with higher radiation dose, somewhat larger treat-ment fields, and higher cumulative doses of cisplatin-based che-motherapy. Thus, both the disease and the treatment can beexpected to have a significant impact on anorectal function andthereby faecal incontinence. Most prostate and cervical cancer pa-tients do not have tumour involvement of the anal- or rectal canal,indicating a causal importance of high dose radiotherapy and notonly tumour destruction [8,9]. Patients treated for prostate cancerin general report lower rates compared to gynaecological- and ano-rectal cancer patients, perhaps because of smaller treatment vol-umes [4,24] while patients irradiated for rectal cancer reporthigh faecal incontinence rates [10], possibly due to the combinedeffect of radiotherapy and surgery.

Leakage of liquid stool was a dominant symptom. Survivorssuffering from diarrhoea may as a consequence be more vulner-able to incontinence [4]. In the current study, only 17% of thesurvivors reported intake of constipating drugs. This might indi-cate under-consumption since most patients with diarrhoea didnot use any constipating drugs. In total, 27% of the survivors re-ported the use of sanitary pads or plugs due to faecal inconti-nence, where of 93% had some degree of faecal leakage.However, 46% of patients with faecal incontinence reported nouse of pad or plug. The low use of pads does not mean that padsare ineffective or not necessary; it may be because many pa-tients are not offered or aware of this remedy. Anal plugs canbe difficult to tolerate [25].

Attempts to treat faecal incontinence are worthwhile. Optimalmanagement requires appropriate investigations and coordinatedmultidisciplinary work as the symptoms and mechanisms arecomplex [26,27]. Although data from treatment of this particularpatient group are sparse, knowledge obtained from other patient-groups could contribute. Conservative approaches such as dietaryadvice, pelvic floor exercises, biofeedback, medication to improvestool consistency, and constipating drugs may be effective[28,29]. Hyperbaric oxygen therapy for late radiation injury indi-cates a positive effect on radiation proctitis although the evidenceis limited [30,31]. Promising data of sacral nerve stimulationsupport further investigation [32]. A stoma may be a preferredsolution in refractory and severe cases of faecal incontinence whenconservative treatment has failed.

60 Faecal incontinence in anal cancer survivors

The current study has limitations. The cross-sectional design ofthe study involves varying lengths of follow-up periods and noability to evaluate the prevalence over time. Elderly and survivorswith comorbidity were more reluctant to participate which mightinfluence the results, even though tumour- and treatment charac-teristics were fairly similar. Regarding the reference group the con-cern for selection bias is greater due to the low response rate,which may have lead to a healthier group with a more positive atti-tude than the actual general Norwegian population. As an attemptto assess whether the current reference group was representativefor the general population, the scores from EORTC QLQ-C30 werecompared to two other sets of normative data. In general, all refer-ence groups followed the same pattern. This is described in detailin a recent published paper [3].

At the same time, the study has several strengths. The studypopulation is unique: a large, unselected cohort, including all dis-ease-free Norwegian patients who received curatively intendedCRT in a seven-year period. Faecal incontinence and HRQOL wereevaluated with validated, well-known and accepted methods thatquantify the patient’s own subjective experience. The outcomeswere compared with symptoms in matched volunteers not treatedfor pelvic cancer.

Norwegian patients treated with curative chemoradiotherapyfor anal cancer have a good prognosis [2]. However, many cancersurvivors suffer from late effects, and there is a need to focus onthe detection and management of debilitating late effects such asfaecal incontinence. Few patients want to present the problemand expose their suffering [33]. There is an established post-treat-ment follow-up, but in addition to examining for recurrence, thisshould include the evaluation of continence as well as other survi-vorship aspects. Despite the complex nature of faecal incontinence,therapeutic possibilities exist that should be considered. Physi-cians should be aware of these and offer treatment and advicefor management as part of the care of anal cancer survivors.

In conclusion, curative chemoradiotherapy for anal cancer usingstandard techniques to a median dose of 54 Gy is associated withhigh score of faecal incontinence in the long-term follow-up.

Conflicts of interest

None declared.

Acknowledgement

This work was supported by research grants from NorthernNorway Health Authority.

References

[1] Glynne-Jones R, Northover JM, Cervantes A, Group EGW. Anal cancer: ESMOclinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol2010;21:v87–92.

[2] Bentzen AG, Guren MG, Wanderas EH, Frykholm G, Tveit KM, Wilsgaard T,et al. Chemoradiotherapy of anal carcinoma: survival and recurrence in anunselected national cohort. Int J Radiat Oncol Biol Phys 2012;83:e173–80.

[3] Bentzen AG, Balteskard L, Wanderås EH, Frykholm G, Wilsgaard T, Dahl O, et al.Impaired health-related quality of life after chemoradiotherapy for analcancer: late effects in a national cohort of 128 survivors. Acta Oncol2013;52:736–44.

[4] Putta S, Andreyev HJ. Faecal incontinence: a late side-effect of pelvicradiotherapy. Clin Oncol (R Coll Radiol) 2005;17:469–77.

[5] Das P, Cantor SB, Parker CL, Zampieri JB, Baschnagel A, Eng C, et al. Long-termquality of life after radiotherapy for the treatment of anal cancer. Cancer2010;116:822–9.

[6] Provencher S, Oehler C, Lavertu S, Jolicoeur M, Fortin B, Donath D. Quality oflife and tumor control after short split-course chemoradiation for anal canalcarcinoma. Radiat Oncol 2010;5:41.

[7] Vordermark D, Sailer M, Flentje M, Thiede A, Kolbl O. Curative-intent radiationtherapy in anal carcinoma: quality of life and sphincter function. RadiotherOncol 1999;52:239–43.

[8] Haugnes HS, Melby B, Larsen KM, Langdal I, Rasi M, Bremnes RM. Assessmentof late urinary, bowel and sexual function after dose escalation from 70 to76 Gy using image-guided radiotherapy in curative treatment of prostatecancer. Scand J Urol Nephrol 2012;46:124–32.

[9] Dunberger G, Lind H, Steineck G, Waldenstrom AC, Nyberg T, Al-Abany M, et al.Self-reported symptoms of faecal incontinence among long-termgynaecological cancer survivors and population-based controls. Eur J Cancer2010;46:606–15.

[10] Bruheim K, Guren MG, Skovlund E, Hjermstad MJ, Dahl O, Frykholm G, et al.Late side effects and quality of life after radiotherapy for rectal cancer. Int JRadiat Oncol Biol Phys 2010;76:1005–11.

[11] Vaizey CJ, Carapeti E, Cahill JA, Kamm MA. Prospective comparison of faecalincontinence grading systems. Gut 1999;44:77–80.

[12] Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, et al. TheEuropean Organization for Research and Treatment of Cancer QLQ-C30: aquality-of-life instrument for use in international clinical trials in oncology. JNatl Cancer Inst 1993;85:365–76.

[13] Maeda Y, Pares D, Norton C, Vaizey CJ, Kamm MA. Does the St. Mark’sincontinence score reflect patients’ perceptions? A review of 390 patients. DisColon Rect 2008;51:436–42.

[14] Fayers P, Aaronson NK, Bjordal K, Groenvold M, Curran D, Bottomley A, et al.The EORTC QLQ-C30 scoring manual. 3rd ed. Brussels: European Organisationfor Research and Treatment of Cancer; 2001.

[15] Macmillan AK, Merrie AE, Marshall RJ, Parry BR. The prevalence of fecalincontinence in community-dwelling adults: a systematic review of theliterature. Dis Colon Rect 2004;47:1341–9.

[16] Nelson R, Norton N, Cautley E, Furner S. Community-based prevalence of analincontinence. JAMA 1995;274:559–61.

[17] Whitehead WE, Borrud L, Goode PS, Meikle S, Mueller ER, Tuteja A, et al. Fecalincontinence in US adults: epidemiology and risk factors. Gastroenterology2009;137:512–7. 517 e511–512.

[18] Rommen K, Schei B, Rydning A, Sultan H, Morkved S. Prevalence of analincontinence among Norwegian women: a cross-sectional study. BMJ Open2012;2.

[19] Perry S, Shaw C, McGrother C, Matthews RJ, Assassa RP, Dallosso H, et al.Prevalence of faecal incontinence in adults aged 40 years or more living in thecommunity. Gut 2002;50:480–4.

[20] Alsadius D, Hedelin M, Johansson K-A, Pettersson N, Wilderäng U, Lundstedt D,et al. Tobacco smoking and long-lasting symptoms from the bowel and theanal-sphincter region after radiotherapy for prostate cancer. Radiother Oncol2011;101:495–501.

[21] Rothbarth J, Bemelman WA, Meijerink WJ, Stiggelbout AM, Zwinderman AH,Buyze-Westerweel ME, et al. What is the impact of fecal incontinence onquality of life? Dis Colon Rect 2001;44:67–71.

[22] Ng M, Leong T, Chander S, Chu J, Kneebone A, Carroll S, et al. AustralasianGastrointestinal Trials Group (AGITG) contouring atlas and planningguidelines for intensity-modulated radiotherapy in anal cancer. Int J RadiatOncol Biol Phys 2012;83:1455–62.

[23] Westbury CB, Yarnold JR. Radiation fibrosis – current clinical and therapeuticperspectives. Clin Oncol 2012;24:657–72.

[24] Olopade FA, Norman A, Blake P, Dearnaley DP, Harrington KJ, Khoo V, et al. Amodified inflammatory Bowel disease questionnaire and the Vaizeyincontinence questionnaire are simple ways to identify patients withsignificant gastrointestinal symptoms after pelvic radiotherapy. Br J Cancer2005;92:1663–70.

[25] Deutekom M, Dobben AC. Plugs for containing faecal incontinence. CochraneDatabase Syst Rev 2012;4:CD005086.

[26] Andreyev HJ, Davidson SE, Gillespie C, Allum WH, Swarbrick E. Practiceguidance on the management of acute and chronic gastrointestinal problemsarising as a result of treatment for cancer. Gut 2012;61:179–92.

[27] Andreyev J. Gastrointestinal symptoms after pelvic radiotherapy: a newunderstanding to improve management of symptomatic patients. LancetOncol 2007;8:1007–17.

[28] Norton C, Whitehead WE, Bliss DZ, Harari D, Lang J. Management of fecalincontinence in adults. Neurourol Urodyn 2010;29:199–206.

[29] Heymen S, Scarlett Y, Jones K, Ringel Y, Drossman D, Whitehead WE.Randomized controlled trial shows biofeedback to be superior to pelvic floorexercises for fecal incontinence. Dis Colon Rect 2009;52:1730–7.

[30] Craighead P, Shea-Budgell MA, Nation J, Esmail R, Evans AW, Parliament M,et al. Hyperbaric oxygen therapy for late radiation tissue injury in gynecologicmalignancies. Curr Oncol 2011;18:220–7.

[31] Clarke RE, Tenorio LM, Hussey JR, Toklu AS, Cone DL, Hinojosa JG, et al.Hyperbaric oxygen treatment of chronic refractory radiation proctitis: arandomized and controlled double-blind crossover trial with long-termfollow-up. Int J Radiat Oncol Biol Phys 2008;72:134–43.

[32] Maeda Y, Høyer M, Lundby L, Buntzen S, Laurberg S. Temporary sacral nervestimulation for faecal incontinence following pelvic radiotherapy. RadiotherOncol 2010;97:108–12.

[33] Johanson JF, Lafferty J. Epidemiology of fecal incontinence: the silent affliction.Am J Gastroenterol 1996;91:33–6.