Enteroscopy and ablation of small bowel angiodysplasia improves quality-of-life and fatigue: A prospective study including biomarker assessment

Abstract

BACKGROUND

Small bowel angiodysplasia (SBA) are an increasingly recognised cause of chronic anaemia and gastrointestinal bleeding. Gastrointestinal angiodysplasia have been independently associated with clinically significant impairments in health-related quality of life (QoL) and fatigue. Balloon assisted enteroscopy (BAE) and argon plasma coagulation (APC) along with general supportive measures are the standard of care, albeit with high relapse rates. The effect of endoscopic therapy on the health related QoL in patients is unclear.

AIM

To assess the impact of BAE and APC on the QoL of patients with SBA and explore potential predictors of response.

METHODS

Symptomatic SBA patients referred for BAE were prospectively recruited from December 2023 to December 2024. Subjects completed a short form-36 health survey (SF-36) and Fatigue Severity Scale (FSS) and provided bloods prior to treatment. Repeat SF-36 and FSS assessments were completed a minimum of 6 months post treatment. Scores pre and post intervention were compared for each subject. Results were controlled for age, gender, haemoglobin level and angiogenic factors.

RESULTS

Of 40 enrolled subjects, 31/36 (86%) who underwent enteroscopy with APC have completed both assessments. There was a statistically significant improvement in both FSS and disease appropriate SF-36 domains over time that was independent of age, gender and haemoglobin level. Of interest, higher endostatin levels were associated with better improvement in fatigue scores.

CONCLUSION

BAE with APC for SBA is beneficial and results in a significant improvement in both patients reported heath related QoL and fatigue severity irrespective of baseline demographics.

Key Words: Small bowel angiodysplasia; Balloon assisted enteroscopy; Quality of life; Angiogenic factors; Fatigue

Core Tip: This prospective study in subjects with small bowel angiodysplasia has demonstrated that endoscopic intervention with balloon assisted enteroscopy and argon plasma coagulation has a positive impact on both patients reported quality of life and fatigue severity. The beneficial affect is independent of age, gender and baseline haemoglobin levels. The study supports the use of balloon assisted enteroscopy and argon plasma coagulation as a cornerstone of small bowel angiodysplasia treatment, particularly for those with multiple lesions and those amenable to intervention, as benefit was demonstrated without adjunctive medical therapy. Endostatin would appear to be a potential biomarker for response and warrants additional investigation.

INTRODUCTION

Balloon assisted enteroscopy (BAE) and argon plasma coagulation (APC) along with general supportive measures are the standard of care, albeit with high anaemia and bleeding recurrence rates for small bowel angiodysplasia (SBA)[1,2]. In general, there is a recognition that patient reported outcomes are an important assessment of quality in all endoscopy including BAE[3]. Similarly, quality of life (QoL) is an important outcome measure to assess the utility of any given treatment strategy particularly in patients with chronic or relapsing gastrointestinal conditions and should be assessed in addition to biochemical and clinical outcomes[4]. Previous research has shown gastrointestinal angiodysplasia are independently associated with fatigue and clinically significant impairments in health-related QoL, which can improve with medical therapy[5,6]. The effect of endoscopic therapy on fatigue and health related QoL in patients with SBA is unclear.

Several angiogenic factors have been linked with symptomatic sporadic angiodysplasia and associated overt and occult bleeding[7-9]. Their role in disease evolution and progression is not fully understood, nor has their utility in clinical practice as biomarkers been fully explored. This study aims to prospectively assess the impact of BAE with APC on the health related QoL and fatigue in a cohort of patients with capsule endoscopy confirmed SBA and to explore potential predictors of response.

MATERIALS AND METHODS

Ethical approval was granted by the St James’s Hospital/Tallaght University Hospital Research Ethics Committee, No. 2020-10 Chairman’s Action (37). Adult patients referred to our centre for device assisted enteroscopy and APC of symptomatic (anaemia or gastrointestinal bleeding) capsule endoscopy confirmed significant (P2 lesions) SBA, were prospectively recruited from June 2023 to December 2024. Patients were invited to participate on the day of their procedure. Following informed consent, subjects completed both validated short form-36 health survey (SF-36) and Fatigue Severity Scale (FSS) questionnaires and provided serum for angiogenic factor analysis before undergoing endoscopy and treatment[10,11]. Patient demographics, base line bloods and endoscopy findings were documented.

Patients with multifocal gastrointestinal angiodysplasia, those with previous device-assisted enteroscopy (DAE) treatments for SBA, those who underwent DAE without intervention (lesion not identified or reached), those who failed endoscopic therapy requiring alternative intervention to control bleeding, those on maintenance medical therapy including somatostatin analogues, thalidomide, anti-vascular endothelial growth factor therapies, tranexamic acid, or those unable to complete either survey, those with concurrent active malignancy, valvular heart disease scheduled for replacement and those with hereditary haemorrhagic telangiectasia were excluded. Patients were contacted by phone 6 months after inclusion and a repeat SF-36 and FSS completed during the call. While need for re-intervention (DAE + APC) was assessed and use of adjunctive medical therapies, specific evidence of recurrence or rebleeding was not sought or available (fecal immunochemical test, haemoglobin, repeat capsules).

All procedures were performed with a balloon assisted enteroscope in accordance with our unit’s protocol and international guidance, the majority under conscious sedation. Findings and intervention including APC watts and pulses, adjunctive therapy (clips) were recorded as normal in the procedure report. Patients and the referring clinician were appraised of the results of the BAE. Specific instructions on maintenance or supportive care was not given. Both monitoring format and intervals and supportive care was left to the discretion of the primary team.

Angiogenic factor analysis

Serum was collected on the day of the procedure as standard and stored for batch analysis. Quantification of angiogenic factors including angiopoietin 1 (Ang-1), angiopoietin 2, vascular endothelial growth factor, endostatin, and tissue inhibitor of metalloproteinases-1 levels were performed using commercially available sandwich enzyme-linked immunosorbent assay kits (R&D Systems, Oxford, United Kingdom), according to the manufacturer’s instructions. Absorbance was measured at 450 nm using a microplate reader. Standard curves were generated and concentrations of angiogenic factors in samples interpolated from the standard curves.

Statistical analysis

SF-36 domain and FSS scores pre and post BAE and intervention (APC) were compared for each subject over time. A FSS of ≥ 4 was considered to represent significant fatigue, while each SF-36 domain was valued from 0-100, with higher scores associated with a more favourable health state. Results were controlled for age, gender, angiogenic factors and haemoglobin level. Initial analysis was conducted with the assistance of artificial intelligence (Microsoft Copilot, WA, United States) with findings confirmed by univariate and multivariate analysis with SPSS V.30 using standard statistical methods, including a paired t-test, Mann-Whitney U, Spearman correlations, ANOVA and linear regression. To determine the required sample size to detect a clinically significant change in QoL parameters (scores 0-100) and FSS (scores 0-7), we performed a standard power calculation to detect a change of 25 points and 2 points prospectively, with the following assumptions, significance level of 0.05, power of 80%, and standard deviations of 20 and 1.5. The estimated sample size was 22 and 16 respectively.

RESULTS

Study population

To date 40 patients have been enrolled and completed baseline SF-36 and FSS questionnaires and provided serum for angiogenic factor analysis. BAE, all but one via an anterograde approach, was complete in all cases, with identification and treatment with APC of angiodysplasia in 36 (90%). Multiple angiodysplasia were treated in the majority 32/36 (89%). APC treatment was provided at 25-30 watts at a flow rate of 0.8 L/minute, the median pulses reported was 12, clips were also applied in three cases. There were no procedural associated complications and no subject required repeat BAE during the study period. To date 31/36 (86%) who underwent enteroscopy and APC have been contacted and supplied repeat questionnaires, three individuals could not be reached and were lost to follow up and 2 had not reached the follow up requirement of 6 months. The median interval to follow up was 7.8 months. For this cohort 19 (61%) were male, the mean age was 72 ± 8.6 years (range 49-84) and mean haemoglobin was 11.1 ± 2.4, range 4-14.6 g/dL (Figure 1).

Figure 1 Study population. SBCE: Small bowel capsule endoscopy; SB: Small bowel; BAE: Balloon-assisted enteroscopy; APC: Argon plasma coagulation.

SF-36 and FSS questionnaire

Baseline and follow up SF-36 domain scores are shown in Table 1. In summary, there was a statistically significant improvement in SF-36 scores from baseline to follow up post BAE and APC in all domains except for role limitation due to emotional difficulties and pain. Similarly, there was a significant reduction in fatigue scores over time from a mean of 4.7 (median 4.9) at base line to 2.5 (median 2) on follow up, with a mean change of -2.2, 95% confidence interval (CI): -2.8 to 1.6, P < 0.001. Age, gender and haemoglobin level did not significantly affect the changes in FSS or SF-36 domains (Table 2).

Table 1 Quality of life domain scores at base line and follow up, mean/median.

Domain	PF	RL-P	RL-E	E/F	MH	SF	Pain	GH
Baseline	44.8/40	23.6/0	50.5/33	34.8/45	65.2/68	63.1/63	55.6/65	45.3/46
Follow up	62.4/65	54.8/50	66.7/100	54/50	84.1/92	90.7/100	61.2/60	63.4/65
Δ Mean (95%CI)	17.5 (8-27)	31.2 (6-57)	16.2 (-8 to 40)	19.2 (8-31)	18.7 (9-29)	27.7 (16-40)	5.6 (-11 to 23)	18 (10-26)
P value	0.003	0.017	0.178	0.002	0.001	0.0001	0.52	0.0001

PF: Physical functioning; RL-P: Role limitation physical; RL-E: Role limitation emotional; E/F: Energy/fatigue; MH: Mental health; SF: Social functioning; GH: General health; CI: Confidence interval.

Table 2 Demographics and haemoglobin impact on change in significant quality of life domains and fatigue severity scale.

Domain	Predictor	Coefficient	P value	CI lower	CI upper	R2d
PF change1	Age	0.496799029	0.5493559	-1.1865686	2.18016665	0.048581685
PF change	Gender	7.609259284	0.5615835	-18.989681	34.2082	0.048581685
PF change	Hb	1.963534859	0.4963046	-3.8858508	7.81292048	0.048581685
RLP change	Age	-0.144349859	0.9149046	-2.894301	2.60560127	0.246642284
RLP change	Gender	36.69853355	0.0943931	-6.7535241	80.1505912	0.246642284
RLP change	Hb	7.170977096	0.1350192	-2.3845854	16.7265396	0.246642284
EF change	Age	0.946169468	0.2363515	-0.6583037	2.55064259	0.059013684
EF change	Gender	-5.636567085	0.6514632	-30.988894	19.7157594	0.059013684
EF change	Hb	1.086581889	0.6919798	-4.4886602	6.66182399	0.059013684
GH change	Age	-0.894963168	0.1269223	-2.0616117	0.27168537	0.101928309
GH change	Gender	5.985909667	0.5103566	-12.448338	24.420157	0.101928309
GH change	Hb	-2.06762978	0.3041048	-6.1215138	1.98625427	0.101928309
MH change	Age	-0.607081002	0.3391156	-1.8884905	0.67432848	0.212102758
MH change	Gender	21.07520033	0.0419478	0.82761193	41.3227887	0.212102758
MH change	Hb	0.489064312	0.823144	-3.9635924	4.94172105	0.212102758
SF change	Age	-0.390959905	0.6266197	-2.0232674	1.24134754	0.135090432
SF change	Gender	20.2213016	0.1191332	-5.5708357	46.0134389	0.135090432
SF change	Hb	1.068574657	0.7017204	-4.6033865	6.74053578	0.135090432
FSS change	Age	0.312632	0.531044	-0.70421	1.329473	0.158169
FSS change	Gender	7.96919	0.30262	-7.66476	23.60314	0.158169
FSS change	Hb	2.272987	0.188219	-1.19423	5.740201	0.158169

¹Difference in score from baseline and follow up.

PF: Physical functioning; RLP: Role limitation physical; EF: Energy fatigue; MH: Mental health; GH: General health; SF: Social functioning; FSS: Fatigue Severity Scale; Hb: Haemoglobin; CI: Confidence interval.

Biomarkers and impact on patient reported outcome

The median/mean values, SD and CIs for each of the five angiogenic factors for the whole cohort are shown in Table 3. There was no statistical association between any of the angiogenic factors and patient characteristics, part from an inverse relationship between haemoglobin level and Ang-1 values (Table 4). Of interest endostatin levels were negatively associated with baseline fatigue scores (r = -0.53, P = 0.004, 95%CI: -0.75 to -0.19) and positively affected the degree of change in FSS scores over time (r = 0.45, P = 0.016, 95%CI: 0.1-0.71; Figure 2). While a higher baseline Ang-1 was associated with less improvement in the role limitation physical domain, r = -0.051, P = 0.005, 95%CI: -0.74 to -0.21, this association did not persist when controlled for haemoglobin.

Figure 2 Endostatin and fatigue severity scale association. FSS: Fatigue Severity Scale; CI: Confidence interval.

Table 3 Angiogenic factor results for the cohort.

Factor (dilution), pg/mL	Mean	Median	SD	95%CI
Endostatin (1/5)	2773.3	2763.9	417.3	2620-2926
Ang-1 (1/2)	598.8	497.9	394.8	435.9-743.64
Ang-2 (1/50)	15231.7	11514.5	11285.32	11092.3-19371.3
VEGF (1/2)	233.8	108.4	349.3	105.7-362.02
Timp-1 (1/500)	107269.3	78567.1	69069.6	81934.4-132604.2

Ang-1: Angiopoietin 1; Ang-2: Angiopoietin 2; VEGF: Vascular endothelial growth factor; Timp-1: Tissue inhibitor of metalloproteinases-1; CI: Confidence interval.

Table 4 Demographics and angiogenic factor association.

Angiogenic factor (dilution), pg/mL	Predictor	Coefficient	P value	R2
Ang-1 (1/2)	Age	-3.9091	0.6688	0.2064
Ang-1 (1/2)	Gender	170.7615	0.2425	0.2064
Ang-1 (1/2)	Hb	-79.3965	0.0178	0.2064
Ang-2 (1/50)	Age	45.1366	0.8759	0.0428
Ang-2 (1/50)	Gender	3435.2961	0.4542	0.0428
Ang-2 (1/50)	Hb	480.0491	0.6333	0.0428
VEGF (1/2)	Age	3.3534	0.6846	0.1569
VEGF (1/2)	Gender	98.5417	0.4517	0.1569
VEGF (1/2)	Hb	48.1964	0.1012	0.1569
Endostatin (1/5)	Age	-2.7524	0.8002	0.0332
Endostatin (1/5)	Gender	66.3708	0.6996	0.0332
Endostatin (1/5)	Hb	20.2682	0.5926	0.0332
Timp1 (1/500)	Age	2216.3375	0.217	0.0677
Timp1 (1/500)	Gender	-17153.681	0.5408	0.0677
Timp1 (1/500)	Hb	3417.3353	0.5793	0.0677

Ang-1: Angiopoietin 1; Ang-2: Angiopoietin 2; VEGF: Vascular endothelial growth factor; Timp-1: Tissue inhibitor of metalloproteinases-1; CI: Confidence interval; Hb: Haemoglobin.

DISCUSSION

Our prospective study is the first to show BAE and intervention with APC for symptomatic SBA has a positive impact on patient reported fatigue and QoL domains relating to general health, physical functioning and role limitation associated with physical problems, as well as mental health and social functioning. Domains relating to pain and role limitation due to emotional problems were not affected, which is in keeping with the nature of the condition. The improvement in QoL parameters post BAE and APC remained significant on controlling for age, gender and baseline haemoglobin levels. Our findings are similar to previous studies assessing medical therapy[6].

The majority of patients included in our study had multiple lesions detected and treated on BAE, which may bias our study in favour of a good response to intervention. However, in our experience most patients presenting with sporadic SBA have multiple lesions in keeping with this cohort and as such our study reflects clinical practice[11]. In addition, we assessed the impact of therapy on QoL and fatigue in patients referred for BAE and intervention only. It is assumed that some patients with isolated, difficult to reach, or less significant, P1 lesions or those with significant comorbidity precluding enteroscopy were not included. As such, our data cannot be extrapolated to suggest all SBA patients would benefit from BAE and APC.

As patients included were referred after a positive capsule endoscopy with known SBA, it is assumed all were already on supportive care, which accounts for the finding of a median baseline haemoglobin level of 11.1 g/dL. Similarly, there was no specific instruction given post BAE and intervention regarding monitoring or ongoing supportive care. As such, the impact of ongoing iron supplements, or transfusions and degree of local monitoring cannot be assessed. Unlike other studies of medical therapy, which allowed endoscopic intervention as required, adjuvant medical treatment including somatostatin anlogues were excluded in our cohort, to allow us to specifically assess the effect of enteroscopy and APC on QoL and FSS. The impact of combined endoscopic and medical therapy is therefore beyond the scope of our paper. Of note haemoglobin level alone should not be taken as a reason to defer BAE and APC, as the majority of patients in our study benefited from endoscopic ablation, irrespective of haemoglobin.

In general, the majority of angiogenic factors previously associated with sporadic SBA did not correlate with changes in either SF-36 scores or FSS scale[12-15]. While the relatively small sample size and lack of control group could account for this, our findings may simply reflect the fact that these factors are altered in response to a systemic driver, linked to SBA pathogenesis, such as hypoperfusion or hypoxia, which is not altered by endoscopic intervention. In addition, ongoing angiogenic abnormalities likely account for the high rate of recurrent disease and need for re-intervention. As such, targeted interventions against these factors should remain an ongoing focus of investigation, in an attempt to alter the natural history of the disease. In contrast to other factors, endostatin, a protein known to have anti-angiogenic effects, and under investigation as a cancer therapy, had a statistically significant effect on fatigue. Our preliminary data suggests higher endostatin levels may be linked with lower baseline and greater improvements in FSS scores, suggesting a possible protective role in our SBA cohort. Our findings suggest endostatin warrants further examination and validation as an independent biomarker of disease and potential treatment target, in larger multicenter translational as well as basic science studies.

The virtual nature of our follow up, a telephone call rather than physical appointment, to accommodate patients referred from long distances to our tertiary referral unit, precluded the collection of follow up bloods. This is a weakness as repeated sampling could have shed further light on the impact of BAE on haemoglobin and on angiogenic factors. Similarly, the sample size is relatively small which limits interpretation, albeit is adequate based on our power calculations to elicit a clinical effect. Isolated SBA is not common and our cohort from a single expert centre, which precluded significant variation in enteroscopy practice, compares favourably with other multicentre studies.

The median time to follow up in our cohort was only 7.8 months which is likely to be too short to evaluate both the long term impact of treatment on QoL as well as disease recurrence. Reported recurrence rates are significant, 80% with a mean time to representation of 10.7 months in a previous cohort of ours[11]. Of note, no one in our cohort needed re-intervention over the study period. It would be of interest to reassess the impact of retreatment with BAE and APC on this cohort.

CONCLUSION

Our study has shown BAE and ablation of SBA significantly improves fatigue and related QoL, independently of demographics and baseline haemoglobin.