A systematic review of symptomatic small bowel lipomas of the jejunum and ileum

Introduction Small bowel lipomas are rarely encountered benign adipose growths found within the small intestine wall or mesentery. Limited up-to-date evidence exists regarding such lipomas. We aim to aid clinical decision-making and improve patient outcomes through this comprehensive review. Methodology The terms ‘small bowel,’ ‘small intestine,’ ‘jejunum’ and ‘ileum’ were combined with ‘lipoma.’ EMBASE, Medline and PubMed database searches were performed. All papers published in English from 01/01/2000-31/12/2019 were included. Simple statistical analysis (t-test, Anova) was performed. Results 142 papers yielded 147 cases (adults = 138, pediatric = 9). Male = 88, female = 59 (average age = 49.9 years). Presenting symptoms: abdominal pain = 68.7%; nausea/vomiting = 35.3%, hematochezia/GI bleeding = 33.3%; anaemia = 10.9%; abdominal distension = 12.2%; constipation = 8.9%; weight loss = 7.5%. Mean preceding symptom length = 58.1 days (symptoms >1 year excluded (n = 9)). Diagnostic imaging utilised: abdominal X-Ray = 33.3%; endoscopy = 46.3%; CT = 78.2%; ultrasound = 23.8%. 124/137 (90.5%) required definitive surgical management (laparotomy = 89, laparoscopcic = 35). 9 patients were successfully managed endoscopically. Lipoma location: ileum = 59.9%, jejunum = 32%, mesentery = 4.8%. Maximal recorded lipoma size ranged 1.2–22 cm. Mean maximum lipoma diameter and management strategy comparison: laparotomy 5.6 cm, laparoscopic = 4.4 cm, endoscopic = 3.7 cm, conservative = 4.5 cm. One-way Anova test, p value = 0.21. Average length of stay (LOS) was 7.4 days (range = 2–30). T-test p value = 0.13 when comparing management modalities and LOS. 4 complications, 0 mortality. Conclusions Important previously undocumented points are illustrated; a clearer symptom profile, diagnostic investigations utilised, size and site of lipomas, types and effectiveness of management modalities, associated morbidity and mortality. Open surgery remains the primary management. No statistically significant difference in LOS and lipoma size is demonstrated between management strategies. Endoscopic and laparoscopic techniques may reduce utilising invasive surgery in the future as skillset and availability improve.


Introduction
Little up to date evidence exists regarding lipomas of the small bowel other than anecdotal case reports. Much of the data quoted by these papers can be traced back to epidemiological studies carried out over 20 years ago. More recent studies relate to reviews of duodenal [1] and colonic [2] lipomas. However, no current systematic review exists pertaining to symptomatic lipomas of the ileum and jejunum, which for the purposes of this paper we shall refer to as small bowel lipomas.
Small bowel lipomas are rarely encountered benign adipose growths found within the wall or mesentery of the small intestine. Incidence of intestinal lipomas ranges from 0.035% to 4.4% [3]. Lipomas can arise throughout the gastrointestinal tract with the small bowel accounting for 25% [4]. These benign tumors arise from the sub mucosa of the small intestine in 90% of cases [5]. Small bowel lipomas are most commonly found incidentally with the majority of patients being asymptomatic. Unlike more proximal and distal lesions that can be easily accessed and investigated with endoscopy, small bowel tumors represent a difficult diagnostic entity. The clinical picture is often not clear, with vague symptoms commonly reported.
Of the 797 papers derived from database and hand searches, 504 titles/abstracts remained once duplicates had been removed. These were screened with a further 210 papers excluded because they were not relevant to the paper. Two independent reviewers then reviewed 294 full-text articles. A further 152 papers were excluded; 103/152 were either abstract only (full text not accessible or published), in a different language or not case specific; 18/152 related to duodenal lipomas; 5/ 152 were incidental lipomas in asymptomatic patients; the remaining 26/152 were unable to be accessed. Thus a total of 142 were included yielding a total of 147 cases.
All papers related to individual cases or case series. Given the observational nature of such reports and that no randomised control trials were included, reporter and publication bias was deemed to be low. No funding or other financial support was received in relation to this study.
Lipoma size was recorded in 115 cases and ranged from 1.2 to 22 cm at the greatest diameter. Mean size was 5.1 cm. When comparing mean lipoma size and successful management strategy, laparotomy = 5.6 cm, laparoscopic = 4.4 cm, endoscopically managed 3.7 cm, conservative 4.5 cm (Table 4. One-way Anova test was performed, the p value of 0.21 demonstrated no statistically significant difference between groups (laparotomy, laparoscopic and endoscopic).
Of the 147 cases, one report described the specimen as a chondrolipoma. All other cases were benign lipomas.
Average length of hospital stay (n = 68) was 7.4 (2-30) days. Interquartile range = 3 (Q3-Q1 ). Average length of stay was 8.5 days with open surgery and 6.4 days with laparoscopic surgery (Table 4). T-test was performed, analysing length of stay between laparotomy and laparoscopic management. A p value of 0.13 demonstrated no statistically significant difference in length of stay. Numbers did not permit comparison of length of stay with the other management modalities. Comparison of lipoma size and length of stay in the 52 cases where both variables were recorded is shown in Fig. 3. There was no significant correlation (R 2 = 0.0074). 4 complications were reported from the 135 cases: 2 surgical wound infections; multi-organ failure and PE; intraoperative laceration to muscular layer. No associated intraoperative or 30-day mortality was reported.

Discussion
Our data identify a male preponderance (60%) in those with symptomatic small bowel lipomas. Lipomas of the colon are reported as being more common in women [8], whereas those found in the oesophagus have a greater prevalence in men [9]. Gastrointestinal lipomas are most commonly found in patients aged 50-70 years [10,11]. Average patient age of this cohort (49.9 years) lies just outside this range (however this is comparable with other reported groups of patients with lipomas). Our Table 2 Presenting symptom profile Lipomas of the gastrointestinal tract have been extensively documented as causative factors in bleeding, intussusception, obstruction, volvulus and altered bowel habit. There is wide variation in presentation. (Our data highlights the breadth of presenting symptoms.) Abdominal pain was the most prevalent symptom, reported in 68% of patients, whilst nausea and/or vomiting and gastrointestinal bleeding were also commonly seen, in 35% and 34% of patients respectively. This is not surprising given that 60% of cases were related to intussusception.
Our results are consistent with data from other papers which identify chronic intermittent cramping abdominal pain associated with nonspecific signs of bowel obstruction including nausea, vomiting, gastrointestinal bleeding, constipation or abdominal distension as key symptoms associated with intussusception [12]. Lipomas accounts for 5% of all cases of intussusception in adults [13], the rest of which are mainly caused by malignant neoplasm [14].
The time course of presenting symptoms ranged from only a few hours to many years. Whilst there was considerable discrepancy in time course within our data, the mean of 58.1 days of preceding symptoms (when 9 results >1 year were excluded) serves as an indicator as to the most commonly encountered presentation. The wide variation may be explained by the fact that many patients had undergone semi-urgent/ elective diagnostic investigations in the community prior to presenting as an emergency.
Gastrointestinal endoscopic investigations are viewed as the gold standard to investigate red flag symptoms of malignancy, bleeding, weight loss, on-going abdominal pain and anaemia [15]. Such symptoms are common to both gastrointestinal malignancy and symptomatic lipomas. However, endoscopic investigations are often negative in lipoma patients given the anatomical position of small bowel lipomas. Thus, delay in diagnosis and referral on for further investigations are likely outcomes.
It is not surprising that the majority of patients in our cohort underwent numerous investigations prior to definitive diagnosis and management. Negative endoscopic investigations were a recurrent theme in many. Given documented colonoscopy perforation rates of 0.016%-0.2% [16], are these patients being exposed to unnecessary risk of potential morbidity? This is pertinent, as radiological imaging is an effective diagnostic tool for lipomas. Nevertheless, malignancy is a differential diagnosis and CT alone may miss a small bowel cancer, diagnosis is only accurate in 55% of cases [17]. Thus, endoscopic work up is an important adjunct helping clinicians exclude other more common pathology despite associated risks.
As stated conventional endoscopic investigations such as colonoscopy and gastroscopy are negative in this cohort. However, double balloon enteroscopy (DBE) appears both an effective diagnostic and therapeutic modality enabling direct visualisation, biopsy and resection of small bowel lipomas in the appropriate setting. The ability to offer therapeutic treatment sets this option out from other diagnostic modalities such as video capsule endoscopy [18]. Nevertheless, DBE is an invasive procedure and is limited to specialist centres. Currently DBE does not appear to form part of standard diagnostic work in this patient cohort.
The sensitivity and specificity of ultrasound in the diagnosis of lipomas are reported as being 85.71% and 95.95% respectively by Rahmani et al. [19]. However, transabdominal ultrasonography is not accurate for detecting small bowel tumors; the reported sensitivity is low (26%) [20]. In contrast, CT and MRI both have high sensitivity in detecting gastrointestinal lipomas [21]. It therefore follows that the majority of patients underwent CT imaging (78%).
More lipomas were located in the ileum than jejunum (59.9%-32%          respectively). Our data support previous reports that ileal lipomas are more prevalent than jejunal lipomas [22,23]. Manouras et al. state 'lesions less than 1 cm are considered incapable of producing symptoms, while 75% of those greater than 4 cm are symptomatic' [4]. Our data support this statement, with the average maximal diameter in symptomatic lipomas measuring 5.1 cm. No lipoma less than 1.2 cm was recorded within our dataset. When evaluating whether any association between maximum lipoma diameter and successful treatment modality exists our results suggest that larger lipomas are more likely to undergo surgery (surgically managed = 5.1 cm, endoscopically managed 3.7 cm, conservative 4.5 cm). Caution when interpreting such results should be taken given the small sample sizes of those managed conservatively and endoscopically.
No reports of surveillance relating to small bowel lipoma growth are reported. One may postulate that even incidentally found large (>2 cm) asymptomatic small bowel lipomas do not require routine follow up given the rarity of patients becoming symptomatic and very low associated risk of malignant transformation.
Various pathophysiological mechanisms are shown in Fig. 2. Some are related, with gastrointestinal bleeding occurring as a result of pressure necrosis and ulceration, and obstruction when a lipoma occludes the bowel lumen. Intussusception and volvulus are similarly capable of causing obstruction and bleeding. Intussusception was the most common pathophysiological mechanism within our cohort. Our data give an up to date review of ways in which lipomas give rise to pathology in these patients.
With few documented cases, no consensus on the management of symptomatic small bowel lipomas currently exists. Parallels can be drawn from the management of colonic lipomas where Nallamothu et al. advocate surgery as first line treatment in lipomas that are sessile, with limited peduncles, extension into muscularis propria/serosa, or when endoscopic resection has failed [8]. Surgery is also suggested as primary management for giant colonic lipomas (>4 cm). However, we suggest other strategies may sometimes have a role.
Conservative management alone was effective in 4 patients. Spontaneous expulsion of a 7 × 4.5 × 3.6 cm ileal lipoma resolved a patient's intussusception and negated the need for surgical intervention as described by Kang [24]. Kim et al. report a 50-year-old man who declined surgery after double balloon enteroscopy diagnosed multiple jejunal lipomatosis [25]. He was treated with analgesia and followed up regularly as an outpatient. Suarez et al. document a 51-year-old male found to have multiple submucosal lipomas in the stomach and small bowel [26]. His symptoms spontaneously settled without the need for any treatment. Nevertheless, these cases appear to be the exception within this cohort.
Endoscopic mucosal resection (as part of DBE) appears to have a limited role in the management of small bowel lipomas. Given the anatomical constraints and required expertise of such procedures this practice is not widespread and accounts for only a small portion of those managed. Nevertheless, successful procedures have been undertaken, as evidenced by our data. Noda et al. report endoscopic mucosal dissection of a 3 cm terminal ileal lipoma [27], whilst Morimoto used a combination of endoscopic snare and IT-knife to perform endomucosal dissection of a 5 cm ileal lipoma although this was complicated by a muscular and serosal layer laceration [28]. Javia reports a patient with a 2 cm terminal ileal lipoma which was excised using endoscopic snare [29]. A patient with a 2 cm lipoma underwent double-balloon-assisted jejunal endoscopic mucosal resection, as reported by Kröner et al. [30]. Such reports demonstrate that both jejunal and ileal lipomas may be managed by endoscopic measures. Only one reported case failed to remove the lipoma, citing the size (3 × 1.5 × 1.5 cm) and wide base as reasons for this. A subsequent laparotomy was required to treat the patient [31]. Careful case selection appears to be an important factor, with some authors stating risks of bleeding and perforation as contraindications to undertaking such procedures [32]. Of the 10 attempted endomucosal resections, 9 were published from 2012 onwards, indicating that this is an emerging area within endoscopy.
Our results show that surgery was the most utilised definitive management strategy. Both open and laparoscopic procedures were undertaken with preponderance for laparotomy as definitive management. Those patients requiring surgery primarily underwent bowel resection and primary anastomosis. Anatomical location determined whether resection was only small bowel or included a portion of large bowel. As Table 4 demonstrates, the average size of symptomatic lipoma resected laparoscopically was (1.2 cm) smaller than those removed via open surgery, however, the exact reasons for this is unclear. Patient selection is likely to a be a factor, with multiple aspects taken into consideration e. g. a surgeon's skillset/standard practice, a centres equipment, critical condition of a patient, degree of bowel obstruction, patient comorbidities and lipoma size. The high rate of surgical management may be attributable to the need to exclude alternative causes for each presentation such as malignancy and the limited practice of alternative management strategies [14,33].
Laparoscopic surgery was unsuccessful in 19% of cases attempted. Authors state a variety of reasons for converting to open surgery. Alsayegh et al. report the use of laparoscopy being diagnostic in a 4 year old before converting to a Pfannenstiel incision in order to resect a 6.7 × 7.6 × 4.4 cm lipoma of the mesentery causing volvulus [34]. Bilgin states that intraoperative adhesions in a case of adult intussusception secondary to a lipoma resulted in conversion [35]. The cost of laparoscopic staplers is highlighted as a factor by Lin for performing a laparoscopy-assisted extracorporeal resection and anastomosis of an intussuscepted segment [36]. Sheehan cites oedema and ischaemia following attempted laparoscopic reduction of an ileocolic intussusception [37].
Associated mortality (0%) and morbidity (2%) rates were low. Given that over 90% of patients underwent surgical intervention in a cohort where average age was almost 50 years, such values are encouraging. However, comparison of morbidity and mortality associated with similar pathologies suggests that complications may have been underreported or not documented. Mortality from adult intussusception increases from 8.7% for benign lesions to 52.4% for those with a malignant cause [38]. Although there are numerous documented cases of gastrointestinal lipomas associated with intussusception, very few report associated morbidity.
Crocetti et al. report an average length of stay in hospital of 5 and 7 days in patients with symptomatic colonic lipomas managed laparoscopically and with open surgery respectively [39]. In our cohort the average length of stay with symptomatic small bowel lipomas was 7.4 days. Open surgery was associated with a longer length of stay (8.5 days) when compared to laparoscopic management (6.4 days). These results are consistent with other reports of shorter hospital stays with laparoscopic management of small bowel obstruction [40].
We acknowledge that there are limitations associated with our study. The paper is based on only those cases documented in the literature. The true incidence of symptomatic lipomas is likely to be higher. Equally we take into account reporting bias. Sub-acute symptomatic patients are unlikely to require emergency intervention and equally will not be reported on. Nevertheless, based on the data collated and analysed we feel able to draw rational conclusions.

Conclusion
We provide a topical and current overview of symptomatic small bowel lipomas. Numerous reports exist of individual cases, referencing small observational studies dating back many decades, but little new data concerning this relatively unknown condition has been collated in recent years. Our study is up to date and practical, presenting new findings, helping provide a framework for classification and management.
A number of important and previously undocumented points are illustrated. A clearer symptom profile is described with most presenting as emergencies necessitating tailored patient care in a timely fashion. Computerised tomography appears to be the primary diagnostic investigation, helping delineate both lipoma and sequelae. Lipomas >1.2 cm may be symptomatic although larger lipomas appear more implicated. Open surgery remains the primary management modality, but smaller symptomatic lipomas may be targeted for laparoscopic surgery in appropriate settings. Laparoscopic surgery is associated with shorter hospitals stays. Endoscopic resection may be a practical first line management in carefully selected patients, although limited data currently exist. Such techniques may reduce the need for invasive surgery in future as skillset and availability improve. Morbidity and mortality rates appear low in this cohort of patients irrespective of lipoma size or management strategy.
We hope that this study offers an insight into the many different facets associated with symptomatic small bowel lipomas. This study adds to the scanty existing knowledge about symptomatic small bowel lipomas. It will inform clinicians and guide management in both the elective and emergency setting to help achieveoptimal patient outcomes.

Provenance and peer review
Not commissioned externally peer reviewed.

Funding
None.

Ethical approval
Not Applicable.

Consent
Not applicable.