This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1. Introduction
2. The clinical picture of DD
3. New understandings in the
pathophysiology of DD
4. Current best practice for DD
5. Conclusion
6. Expert opinion
Review
New physiopathological andtherapeutic approaches todiverticular disease: an updateAntonio TursiGastroenterology Service, ASL BAT, Andria (BT), Italy
Introduction: Diverticular disease (DD) of the colon is a widespread disease,
which shows worldwide increasing incidence and represents a significant bur-
den for National Health Systems. The current guidelines claim that symptom-
atic uncomplicated DD (SUDD) has to be treated with spasmolithics and high-
fiber diet, whereas both uncomplicated and complicated acute diverticulitis
has to be treated with antibiotics. However, new physiopathological knowl-
edge suggests that further treatment may be promising.
Areas covered: Pathogenetic and treatment studies on SUDD and acute
diverticulitis published in PubMed, www.clinicaltrials.gov, and in the main
International Congress were reviewed.
Expert opinion: Although absorbable antibiotics and 5-aminosalycilic acid
seem to be effective in treating SUDD, their role in preventing diverticulitis
recurrence is still under debate. Antibiotic use in managing acute diverticulitis
is at least questionable, and use of probiotics seems to be promising but need
further robust studies to confirm the preliminary results.
Diverticulosis is a structural alteration of the colonic wall characterized by thepresence of pockets (diverticula) that form when colonic mucosa and submucosaherniate through defects in the muscle layer of the colon wall [1]. For many yearsit has been thought that diverticulosis affected exclusively the westernized world.However, recent data showed that the prevalence of colonic diverticulosis is increas-ing throughout the world [2,3]. Although most people with colonic diverticulosisremain asymptomatic, about one-fourth of the patients will develop diverticulardisease (DD) and, of these, 15% will ultimately develop complications [4,5].
Current accepted etiology considers that diverticular formation occurs inwesternized societies due to a lack of fiber in the diet [6,7]. The assumption is thatdecreased dietary fiber intake results in decreased intestinal contents and, hence, adecreased size of the lumen, leading to the transmission of muscular contractionpressure to the wall of the colon rather than to the contents of the lumen. The resultof increased pressure on the wall is the formation of diverticula at the weakest pointin the wall: the sites of penetration by blood vessels.
The process through which a diverticulum becomes inflamed has been linked tothat which causes appendicitis: fecal stasis causes obstruction of the diverticularneck, ultimately leading to perforation of the diverticulum [6].
When diverticulosis becomes symptomatic, it is called DD,which describes the symptoms related to the anatomical--structural changes in the colon that harbors diverticula. Inthe last years, some attempts to improve clinical classificationof DD have been performed [8,9]. However, the current classi-fication of DD is still based on EASGE criteria of 1999,which subdivided DD as symptomatic uncomplicated disease,recurrent symptomatic disease and complicated disease [10].Symptomatic uncomplicated DD (SUDD) is characterized
by nonspecific attacks of abdominal pain without evidence ofan inflammatory process. This pain is typically colicky innature, but can be steady, and is often relieved bypassing flatusor having a bowel movement. Bloating and changes of bowelhabits also can occur due to bacterial overgrowth, and consti-pation is more common than diarrhea. Fullness or tendernessin the left lower quadrant, or occasionally a tender palpableloop of sigmoid colon, is often discovered onphysical examination.Recurrent symptomatic disease is associated with the
recurrence of the symptoms described above, and it may occurseveral times per year.These symptoms may resemble irritable bowel syndrome
(IBS). However, we know today that left lower quadrantpain lasting > 24 h and increased fecal calprotectin expressionare specific for SUDD but not for IBS [11-13]. Hence, the useof these clinical/laboratory criteria may be useful to pose acorrect differential diagnosis.Complicated disease describes all complications related to
DD: hemorrhage, abscess, phlegmon, perforation, purulentand fecal peritonitis, stricture, fistula and small bowel obstruc-tion. Acute diverticulitis (namely, DD with the signs andsymptoms of diverticular inflammation) is included in the
complicated disease. Although a clinical classification of acutediverticulitis is not validated yet, CT scan is able to differen-tiate between uncomplicated (acute diverticular inflammationwithout stenoses and/or fistulas and/or abscesses) and compli-cated diverticulitis (acute diverticular inflammation with ste-noses and/or fistulas and/or abscesses) [14,15].
Today, the widespread use of colonoscopy in assessingabdominal symptoms has increased the incidence of theso-called ‘uncomplicated’ diverticulitis, which is morefrequently diagnosed by endoscopy than would previouslyhave been thought likely. In fact, the percentage of findingsconsistent with diverticulitis ranges from 0.48 to 1.5 of allpatients undergoing colonoscopy, even if without signs orsymptoms suspected for acute diverticulitis [16,17]. There is arisk of colonic perforation in inflamed diverticula foundduring colonoscopy. However, a gentle colonoscopy withminimal air inflation can be carried out safely in thesepatients. If a diagnosis of diverticulitis is posed duringcolonoscopy, the procedure may be terminated, with a verylow risk of perforation [17,18].
Symptomatic uncomplicated and recurrent symptomaticforms of colonic DD generally require medical treatment.Uncomplicated diverticulitis is also generally treated medi-cally, but complicated diverticulitis generally requires surgicaltreatment [1].
3. New understandings in thepathophysiology of DD
Recent observations suggest an understanding of the naturalhistory of DD as a chronic inflammatory bowel disease(IBD). Five findings seem to support this hypothesis:
1) DD shows a significant microscopic inflammatoryinfiltrate [19-22]. This microscopic inflammation, rang-ing from increased chronic lymphocytic infiltration toactive neutrophilic infiltrate, seems to be linked to theseverity of the disease [20-22].
2) DD shows an enhanced expression of pro-inflammatory cytokines as TNF-a [23-25]. Moreover,this cytokines overexpression decreases parallel toresponse to therapy [26,27].
3) Obesity is a risk factor for diverticulitis recurrence dueto the pro-inflammatory effect of the adipokines andchemokines [28].
4) Both persisting endoscopic and histological inflamma-tion has been recently identified as significant risk fac-tors for diverticulitis recurrence [29].
5) Up to 20% of patients complaint for persistent abdom-inal pain after surgical treatment of diverticulitis andquality of life of that patients are significantly worseafter surgery [30]. It has been hypothesizedthat persistent symptoms are linked to increased neuro-peptides in mucosal biopsies, which may reflect
Article highlights.
. Diverticular disease (DD) of the colon is a widespreaddisease, which shows worldwide increasing incidence.
. New pathophysiology hypotheses have underlined therole of inflammation in the occurrence of the disease.
. Nonabsorbable antibiotics and 5-aminosalycilic acidseem to be effective in treating symptomaticuncomplicated DD, but only 5-aminosalycilic acid seemsto be effective in preventing occurrence of acutediverticulitis in those patients.
. The role of nonabsorbable antibiotics and5-aminosalycilic acid in preventing diverticulitisrecurrence is still under debate.
. Despite their large use, managing acute diverticulitis byantibiotics is at least questionable.
. Despite probiotics seem promising in treating thisdisease, further robust confirmative studies are neededto confirm the preliminary results.
This box summarizes the key points contained in the article.
CALPROTECTINA FECAL: Usada para procesos inflamatorios intestinales como EII.
EDSON
Rectángulo
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
EDSON
Resaltado
resolved prior acute inflammation and persistentchronic inflammation [31].
Hence, DD may be considered as a chronic inflammatoryprocess ranging from low-grade inflammation that is localisedwithin the colonic mucosa to a full-blown acute diverticulitisresulting expanding inflammation to the colonic wall due tomicro/macro-perforation of the wall from the diverticulum.Why this occurs is not well understood yet. It has been hypoth-esized that DD causes colonic microbial imbalance, and thatchronic inflammation occurs as a consequence of this imbal-ance [32,33]. According to this hypothesis, it has been hypothe-sized that modulating the microflora or administering anti-inflammatory agents may help to treat DD, to prevent divertic-ulitis, and to reduce diverticulitis recurrence (Table 1).
Treatment studies on SUDD and acute diverticulitispublished in the last 5 years in PubMed, www.clinicaltrials.gov, and in the main International Congress were reviewed.Only data with conclusive results were reported, but the cur-rent ongoing trials of interest for this review were also cited.Studies discussed in the article are reported in Table 1.
4. Current best practice for DD
4.1 Symptomatic uncomplicated DD4.1.1 FiberAccording to the current World Gastroenterology Organiza-tion guidelines, many clinicians advise spasmolithics and ahigh-fiber diet or fiber supplementation, which still representthe first-line treatment for SUDD [34]. However, a recentsystematic review found that high-quality evidence for ahigh-fiber diet in the treatment of DD is lacking, and mostrecommendations are based on inconsistent level 2 and mostlylevel 3 evidence [35]. Nevertheless, high-fiber diet is stillrecommended.
4.1.2 AntibioticsSince 1992, the use of rifaximin has been investigated in thetreatment of SUDD. This is a poorly absorbable antibiotichaving a broad spectrum of action, including againstGram-positive and Gram-negative bacteria, aerobes andanaerobes [36]. It has been successfully used in recent years inthe treatment of SUDD [37] and also seems to be effective inmaintaining SUDD remission [38]. A recent meta-analysis ana-lyzed four prospective randomized trials (only one conductedin double-blind placebo-controlled fashion) including 1660patients. The pooled rate of difference for symptom reliefwas 29.0% in favor of rifaximin (rifaximin vs control; 95%CI 24.5 -- 33.6%; p < 0.0001) with a clinically significantnumber needed to treat (NNT = 3) [39].
4.1.3 MesalazineControlling inflammation by mesalazine is another choice totreat SUDD. Although this drug is effectively used for manyyears in treating IBD, the mechanisms of action are not yet
well understood. Mesalazine acts in the gastrointestinal epi-thelium, through N-Ac-5ASA, the active metabolite of5-ASA (mesalazine), but the molecular mechanisms of itsaction are not clear. It is thought that mesalazine inhibitssome key factors of the inflammatory cascade (cyclo-oxygen-ase, thromboxane synthetase and PAF synthetase), inhibitsthe production of interleukin-1 and free radicals, and hasgot an intrinsic antioxidant activity [40]. In light of the newdata on the role of inflammation in the pathogenesis ofSUDD, it was inevitable that researchers tried to apply mesa-lazine on such avenue.
The favorable effect of mesalazine on SUDD has beendemonstrated by several open-label studies [41]. Moreover,two recent, double-blind, placebo-controlled studiesconfirmed these preliminary results.
The first trial was performed at 17 centers in Germany.The primary end point of the study was to investigate theefficacy and safety of mesalazine granules 3 g/daily versusplacebo in patients with lower abdominal pain as a symptomof SUDD. The primary efficacy end point was change inlower abdominal pain to week 4 (baseline defined usingpain score from 7 days pretreatment). Median change inlower abdominal pain with mesalazine versus placebo was-37 versus -33 (p = 0.374) in the intent-to-treat (ITT) popu-lation, and -41 versus -33 (p = 0.053) in the per-protocol (PP)population. Post hoc adjustment for confounding factors (e.g., ‘baseline pain intensity’, ‘baseline symptom score’, and‘localization of diverticula in the descending colon’) resultedin p = 0.111 (ITT) and p = 0.005 (PP). Between-group differ-ences increased using pain score on day 1 as baseline andreached significance for the PP population (mesalazine -42,placebo -26, p = 0.010). Median change in combinedsymptom score from baseline to week 4 was 257 mm withmesalazine versus 198 mm with placebo (p = 0.064). Moreplacebo-treated patients received analgesic/spasmolytic con-comitant medication (34.4 vs mesalazine 21.4%), indicatingimproved pain relief with mesalazine (p = 0.119). Safety wascomparable [42].
The second multicenter, double-blind, placebo-controlledtrial was recently conducted in Italy. Two hundred and tenpatients were randomly enrolled in a double-blind fashionin four groups: group M (active mesalazine 1.6 g/day plusLactobacillus casei subsp. DG placebo), group L (active L. caseisubsp. DG 24 billion/day plus mesalazine placebo), groupLM (active L. casei subsp. DG 24 billion/day plus active mesa-lazine) and group P (L. casei subsp. DG placebo plus mesala-zine placebo). Patients received treatment for 10 days/monthfor 12 months. Recurrence of SUDD was defined as thereappearance of abdominal pain during follow-up, scoredas ‡ 5 (0, best; 10, worst) for at least 24 h. SUDD recurredin no (0%) patient in group LM, in 7 (13.7%) patients ingroup M, in 8 (14.5%) patients in group L, and in23 (46.0%) patients in group P (LM group vs M group,p = 0.015; LM group vs L group, p = 0.011; LM group vsP group, p = 0.000; M group vs P group, p = 0.000; L group
Prevention of diverticulitis recurrencePrevention of surgery
p = 0.044p = 0.001p = 0.020
Raskin [77] MMX (Lialda� 1.2 vs 2.4 vs 4.8 g/day for 2 years) Prevention of diverticulitis recurrence p = n.s.Kamm [78] MMX (Lialda� 1.2 vs 2.4 vs 4.8 g/day for 2 years) Prevention of diverticulitis recurrence p = n.s.Stollman [75] Eudragit L (Asacol� 2.4 g/day � 3 months) Symptoms control after diverticulitis
Prevention of diverticulitis recurrencep = 0.03p = n.s.
Kruis [79] Granules (Granustix� 3 g/day � 1 year) Prevention of diverticulitis occurrenceControl of pain
Prevention of diverticulitis recurrenceImprovement of quality of life
p = n.s.p = 0.021
Smith [26] Granules (Granustix� 3 g/day � 2 months) Pain control in SUDD p = 0.0413Kruis [42] Granules (Granustix� 500 mg/day � 30 days) Pain control in SUDD p = 0.03Tursi [43] Eudragit L
vs P group, p = 0.000). No adverse events were recordedduring the study [43].
4.1.4 ProbioticsUsing probiotics is a third choice to tread SUDD. Probioticsare living microorganisms that can exert host healthbenefits beyond those of inherited basic nutrition [44]. Physio-pathological action of probiotics includes pathogen adherenceinhibition, increasing IgA secretion in Peyer’s patches, increas-ing the immune system activity inhibiting the release ofanti-inflammatory cytokine, and inhibiting pro-inflammatorycytokines [44]. Moreover, these ‘good’ bacteria may interferewith pathogen metabolism by a mechanism of metaboliccompetition [44].
It is generally thought that DD is affected by bacterial over-growth in the diverticula and that this may cause ischemicphenomena, diverticular and peri-diverticular inflammation,increased exposure to intraluminal antigens and toxins, andbacterial flora changes related to stasis [45]. Therefore, a thera-peutic manipulation of the colonic flora may be useful in con-trolling colonic inflammation and finally in controllingsymptoms in those patients. Widespread use of antibioticscan control bacterial overgrowth and prevent translocationof gut bacteria. On the other hand, some bacteria may providespecific health benefits when consumed as a food componentor in the form of specific preparations of viable microorgan-isms, without the risk of antibiotic resistance. Taking thisinto consideration, recent studies investigated the effect ofprobiotics on the course of SUDD. In 2006, our group con-ducted a multicenter, prospective, randomized, open-labelstudy in order to maintain SUDD remission. Ninety
consecutive patients were enrolled in a 12-month follow-upand treated with mesalazine 1.6 g/day (group M), L. caseisubsp. DG 16 billion/day for 15 days/month (group L); mesa-lazine 1.6 g/day plus L. casei DG 16 billion/day for 15 days/month (group LM). Focusing the attention on group L andLM, we found that 23/29 of group L (on ITT: 76.7%[95% CI: 61.5 -- 91.8]), and 29/29 of group LM (on ITT:96% [95% CI: 94.2 -- 100]) were symptom-free at the endof the follow-up (p < 0.05) [46]. Patients completing this trialwere further followed in a prospective, dose-finding studyconducted in an open fashion: mesalazine 800 mg/daily(group M1) or mesalazine 1.6 g 10 days/month (groupM2); mesalazine 800 mg/daily + L. casei DG 16 billion/dayfor 10 days/month (group LM1) or mesalazine 1.6 g + L. caseiDG 16 billion/day for 10 days/month (group LM2); L. caseiDG 16 billion/day for 10 days/month (group L). No differen-ces were found in maintaining SUDD remission, but,significantly, we found that symptoms recurred in allpatients suspending treatment, half of them developing acutediverticulitis [47].
Annibale et al. randomly treated SUDD patients with highfiber alone (group A, 16 patients), or with twice daily 1probiotic sachet (each containing 12.5 billion of Lactobacillusparacasei subsp. paracasei F19 strain) for 14 days/month for6 months + high-fiber diet (group B, 18 patients), or group Ctwice daily two probiotic sachets + high-fiber diet (group C).The primary end point was a decrease in abdominal pain andbloating intensity after treatment. Bloating decreased signifi-cantly in groups B and C (visual analogue scale [VAS] scoregroup B: 4.6 ± 2.6 vs 2.3 ± 2.0, p < 0.05; group C: 3.9 ±2.9 vs 1.8 ± 2.1, p < 0.05). A decrease in abdominal pain within
Table 1. Studies discussed in the article (continued).
Author (ref.) Bacterial strain End points Results
ProbioticsGiaccari [81] Lactobacillus sp. Improvement of
24 h in these groups did not reach statistical significance. Dur-ing treatment, abdominal pain > 24 h recurred only in group A(p = 0.016), whereas no significant difference regarding abdom-inal pain < 24 h and bloating was observed between the twogroups of patients treated with a low or high probiotic dose [48].More recently, the same group conducted a multicenter,
6-month randomized, controlled, parallel-group interventionwith a preceding 4-week washout period. SUDD patientswere randomized to treatment A (24 patients receiving onesymbiotic sachet containing 5 billion of L. paracasei B21060strain once daily plus high-fiber diet for 6 months) ortreatment B (21 patients receiving high-fiber diet alone for6 months). The primary end point was regression of abdom-inal symptoms and change of symptom severity after 3 and6 months of treatment. In group A, the proportion of patientswith abdominal pain > 24 h decreased from 60 to 20%, then5% after 3 and 6 months, respectively, in group A (p < 0.001)and from 33.3 to 9.5% at both 3 and 6 months in group B(p = 0.03). After 6 months of treatment, the mean VAS valuesof both short-lasting (< 24 h) abdominal pain (VAS,mean ± SD, group A: 4.6 ± 2.1 vs 2.2 ± 0.8, p = 0.02; groupB: 4.6 ± 2.9 vs 2.0 ± 1.9, p = 0.03) and abdominal bloating(VAS, mean ± SD, group A: 5.3 ± 2.2 vs 3.0 ± 1.7,p = 0.005; group B: 5.3 ± 3.2 vs 2.3 ± 1.9, p = 0.006)decreased in both groups, whereas the VAS values ofprolonged (> 24 h) abdominal pain decreased in group A,but remained unchanged in the high-fiber diet group (VAS,mean ± SD, group A: 6.5 ± 1.5 vs 4.5 ± 2.1, p = 0.052; groupB: 4.5 ± 3.8 vs 5.5 ± 3.5) [49].Finally, effectiveness of probiotic treatment has been
confirmed by the recent multicenter, double-blind, placebo-controlled trial already mentioned [43]. In this study, theassociation of mesalazine plus L. casei subsp. DG (groupLM) or L. casei subsp. DG alone (group L) was significantlybetter than placebo in preventing SUDD recurrence (LMgroup vs P group, p = 0.000; L group vs P group,p = 0.000) [43].
4.2 Acute diverticulitisFew high-quality randomized trials, systematic reviews ormeta-analyses have been published on the treatment of acutediverticulitis. Despite the lack of evidence from these high-quality studies, there is some evidence to support approachesto treatment.The ‘blind-pouch’ theory, which implicates fecal stasis and
bacterial overgrowth in the pathogenesis of such conditions asappendicitis, has been similarly used to explain the developmentof diverticular inflammation in diverticulosis [50].In the vast majority of cases, inflammation in diverticulitis is
mild. These patients, affected by uncomplicated diverticulitis,are generally treated with a clear liquid diet and antibiotics asoutpatients [34]. In outpatients, broad-spectrum antibiotics areusually given for 7 -- 10 days. Various antibiotics may be usedin the treatment of acute diverticulitis, ranging from ampicillinto third-generation cephalosporins [34,50-53], ensuring complete
coverage against Gram-positive and Gram-negative, andaerobic--anaerobic bacterial strains [34,50-53].
The combination of ciprofloxacin and metronidazole is acommonly used treatment for uncomplicated diverticulitis[34,50-53], both intravenously and orally, but sometimes theseantibiotics may be poorly tolerated by some patients becauseof their high systemic absorption. According to the AmericanSociety of Colon and Rectal Surgeons (ASCRS), ampicil-lin--sulbactam is a good option in this group of patients [52].
If opioid analgesics are required for pain control, meperi-dine is the preferred option as morphine causes colonic spasmand may accentuate colonic hypersegmentation [34,45,51,53].
Outpatient treatment is effective in most cases, and < 10%of them are readmitted at the emergency room for diverticu-litis within 60 days of the initial evaluation [34,45,51,53].
Hospitalization, with intravenous antibiotic treatment, isusually recommended by current guidelines [34,45,51,53] if thepatient is unable to take oral therapy, is affected by severecomorbidity, fails to improve with outpatient therapy, or isaffected by complicated diverticulitis [34,45,51,53].
Clinical improvement in patients affected by acute divertic-ulitis is generally observed within 3 -- 4 days. If patients areadmitted to hospital, a 7- to 10-day course of oral antibioticsis usually given following discharge [34,45,51,53].
A recent, retrospective study evaluated how many patientsat first episode of acute diverticulitis could be managed asoutpatients [54]. The diagnosis of acute diverticulitis wasconfirmed by computed tomography (CT). The end pointsincluded length of stay, need for surgery, percutaneous drain-age, and mortality. Patients were considered to have had aminimal hospitalization, defined as survival to discharge with-out needing a procedure, hospitalization of £ 3 days and noreadmission for diverticulitis within 30 days after discharge.On a cohort of 639 patients, 368 (57.6%) had a minimal hos-pitalization. Female gender and CT scan findings of free air/fluid were negatively associated with the likelihood of minimalhospitalization. The presence of an abscess < 3 cm and strand-ing on CT did not predict the need for a higher level of care.Unfortunately, the authors failed to identify patients likely toneed only minimal hospitalization, and only free air/liquid ina patient admitted for acute diverticulitis indicates a moresevere clinical course [54].
As stated, few high-quality randomized trials have beenpublished in this field. In 2010, a clinical trial compared theefficacy of short-term therapy (4 days) versus standard therapy(7 days) for uncomplicated sigmoid diverticulitis using ertape-nem 1 g daily. Both patient groups were monitored until dis-charge and were followed up after 4 -- 6 weeks and 52 months.No significant differences were discerned between the twogroups in terms of the basic data, apart from the meannumber of diverticulitis episodes (short term 1.28 ± 0.64 vsstandard 1.64 ± 1.07, p = 0.037). The mean hospital staywas 8.8 days, significantly lower in the short-term groupthan in the standard therapy group (7.8 ± 2.8 vs 9.7 ±3.2 days; p = 0.002). After 4 days, treatment was classified
as having proved successful in 98.0% of cases and after 7 daysin 98.2% of cases. An overall success rate of 95.1% (94.0 vs96.2%, n.s.) was recorded after 1 month [55].
Another recent prospective study compared the efficacy,safety, and costs of hospital treatment with intravenous antibi-otics and outpatient treatment with oral antibiotics. Seventy-six patients were included in the study, of which forty-fourunderwent intravenous treatment with metronidazole500 mg three times daily plus ciprofloxacin 400 mg two timesdaily (hospital treatment group) and thirty-two took oral anti-biotics metronidazole 500 mg three times daily and ciproflox-acin 500 mg two times daily outpatient (group). Outpatienttreatment was viable in almost 95% of those patients sufferingfrom uncomplicated acute diverticulitis as only two patients(6%) required hospital admission after outpatient treatment.A similar rate of complication and relapse was found betweento those patients admitted to hospital and treated with intra-venous antibiotics or treated as outpatients with oral antibiot-ics (p = 0.86). Finally, outpatient approach was able to save~e1600 per patient (p < 0.05) [56].
A recent trial compared inpatient (group 1) versus outpa-tient (group 2) management of uncomplicated left colonicdiverticulitis and to analyze differences in quality of life andeconomic costs. The first dose of antibiotic was given intrave-nously to all patients in the emergency department and thengroup 1 patients were hospitalized, whereas patients in group2 were discharged. The primary end point was the treatmentfailure rate of the outpatient protocol and the need for hospitaladmission. The secondary end points included quality-of-lifeassessment and evaluation of costs. A total of 132 patientswere randomized, and 4 patients in group 1 and 3 patients ingroup 2 presented treatment failure without differencesbetween the groups (p = 0.619). The overall healthcare costper episode was three times lower in group 2, with savings ofe1124.70 per patient. No differences were observed betweenthe groups in terms of quality of life [57].
4.2.1 Are antibiotics always mandatory in acute
diverticulitis?These trials confirm that outpatient treatment is safe andeffective in selected patients with uncomplicated acute diver-ticulitis. Outpatient treatment allows important costs savingto the health systems without a negative influence on the qual-ity of life of patients with uncomplicated diverticulitis reduceshealthcare costs by > 60%.
Unfortunately, evidence of antibiotic use in acute divertic-ulitis, uncomplicated disease included, is not ‘evidence-based’.In this way, two recent randomized studies found antibiotictreatment not superior to simple support therapy in obtainingclinical resolution and preventing diverticulitis recurrence.
A recent retrospective, case--control study in 272 patientswith mild colonic diverticulitis admitted to two hospitalswith distinctly different treatment regimes concerning anti-biotic use was conducted. A total of 191 patients weretreated without antibiotics and 81 with antibiotics. Groups
were comparable at baseline with respect to age, sex, comor-bidity, use of nonsteroid anti-inflammatory drugs, steroidsand aspirin, C-reactive protein, and white blood countlevels. In the antibiotics group, there were significantlymore patients with a temperature of 38.5�C or higher onadmission (8 vs 19%; p = 0.014). Treatment failure didnot differ between groups (4 vs 6%; p = 0.350). The riskof recurrence was higher in the antibiotics group on logisticregression analysis but did not reach statistical significance(OR 2.04; 95% CI 0.88 -- 4.75; p = 0.880). The only factorthat increased the risk of recurrence was nonsteroid anti-inflammatory drug use (OR 7.25; 95% CI 1.22 -- 46.88;p = 0.037) [58].
A multicenter, randomized trial involving 10 surgicaldepartments in Sweden and 1 in Iceland recruited623 patients with CT-confirmed acute uncomplicated left-sided diverticulitis. Patients were randomized to treatmentwith (314 patients) or without (309 patients) antibiotics.Age, sex, body mass index, comorbidities, body tempera-ture, white blood cell count, and C-reactive protein levelon admission were similar in the two groups. Complica-tions, such as perforation or abscess formation, were foundin six patients (1.9%) who received no antibiotics and inthree patients (1.0%) who were treated with antibiotics(p = 0.302). The median hospital stay was 3 days in bothgroups. Recurrent diverticulitis necessitating readmissionto hospital at the 1-year follow-up was similar in the twogroups (16%, p = 0.881) [59].
These trials show that antibiotic treatment for acuteuncomplicated diverticulitis neither accelerates recovery norprevents complications or recurrence, and that it should bereserved for the treatment of complicated diverticulitis. Whythis occurs is unknown, and the use of antibiotics in treatingboth SUDD and acute diverticulitis is becoming question-able: some interesting trials are currently ongoing in orderto answer these questions [60,61].
Another point requiring attention is the role of personalizedmedicine in terms of factors related to genetic background ofthose patients to support such efforts in antibiotic therapy.Individualization of drug therapy, described as tailoring drugselection and drug dosing to a given patient, is an importantobjective in antibiotic therapy in order to optimize the efficacyof a drug, minimize its toxicity, or both. Many specific pointsmay be developed for acute diverticulitis patients, rangingfrom selecting an antibiotic based on minimum effectiveconcentrations and bacterial sensitivity, to assessing pharmaco-kinetics and pharmacogenomics of antibiotics in this specificpopulation. Future clinical trials investigating antibiotictherapy in acute diverticulitis patients have to provideadditional strategies to achieve individualization.
4.2.2 Failing outpatient treatment of uncomplicated
diverticulitisOutpatient treatment is effective in most cases, with a low riskof emergency room readmission for diverticulitis within
60 days of the initial evaluation (< 10%) [1,6]. On thecontrary, the recurrence rate in outpatients is quite higher,occurring up to 18% at 10 years [62].Hospitalization, with intravenous antibiotic treatment, is
usually recommended by current guidelines [34,53] if thepatient is unable to take oral therapy, is affected by severecomorbidity, or the patient fails to improve with outpatienttherapy [34,53].Clinical improvement in patients affected by acute divertic-
ulitis is generally observed within 3 -- 4 days. If patients areadmitted to hospital, a 7- to 10-day course of oral antibioticsis usually given following discharge [34,45,51,53].
4.2.3 Unconventional treatments for uncomplicated
diverticulitisA recent retrospective open-label trial established the role ofJapanese herbal (Kampo) medicine, daiobotanpito (DBT) orDa Huang Mu Dan Tang in Chinese, in the treatment ofacute diverticulitis [63]. DBT has been traditionally used intradition in Kampo medicine for the treatment of abscessesof the intestine, such as diverticulitis or appendicitis [64].Patients took intravenous antibiotics plus DBT (group 1) orwithout DBT (group 2). DBT supplementation was signifi-cantly better in reducing the duration of fever (p < 0.05),abdominal pain (p < 0.05), and antibiotic administration(p < 0.05). A trend toward a shorter hospital stay(p = 0.061) and fasting (p = 0.055) was found, althoughthis did not reach statistical significance.Although limited by the open-label and retrospective
design, this study opens an interesting debated on the use oftraditional herbal medicine as support to standard medicaltreatment.
4.3 Current management of diverticulitis after an
acute episodeAlthough recurrence of diverticulitis is common, there is littleevidence to define the optimal approach to managing divertic-ulitis after an acute episode. We know that the long-termrecurrence rate of diverticulitis is up to 20% [34,45,51,53], evenif a more recent, colonoscopy-based study hypothesized lowerrate of diverticulitis occurrence [65].
4.3.1 High-fiber dietOnce the acute episode has resolved, patients are generallyadvised to maintain a high-fiber diet in order to optimize theirbowel movements [53]. However, the collective literatureinvestigating the role of dietary modification in preventingDD or a recurrence of diverticulitis is inconsistent. Lookingat more recent studies, the results are conflicting [66,67], andthere is no consistent support for recommending a high-fiberdiet. Despite this lack of evidence, a high-fiber diet is stillcommonly recommended to reduce the likelihood of divertic-ulitis recurrence [35].Another interesting point is related to the classical advice to
avoid consuming seeds, popcorn, and nuts, which is based on
the assumption that such substances could theoretically enter,block, or irritate a diverticulum and result in diverticulitis,and possibly increase the risk of perforation. There is,however, no evidence to date to support this practice [68].
4.3.2 AntibioticsGiven the potential involvement of microbial imbalance in thepathogenesis of DD [33], one option to prevent recurrence afteran acute episode may be to use a single, broad-spectrum anti-biotic that has activity against both Gram-negative and anaer-obic bacteria. Recent open-label pilot studies found that cyclicadministration of rifaximin (800 mg/day for 10 days everymonth) can effectively improve symptoms, whereas preventionof acute diverticulitis was not clearly demonstrated [69,70].However, the results of two recent systematic reviews in assess-ing the role of rifaximin in preventing recurrence of diverticu-litis are conflicting [71,72].
4.3.3 SurgerySurgery is considered a therapeutic option after attacksof diverticulitis.
According to the ASCRS guidelines [51] and others [73],elective resection should be considered after one or twowell-documented attacks of diverticulitis, depending on theseverity of the attack and age and medical fitness of thepatient. Although the recurrence rate of diverticulitis after sur-gery is currently considered quite low (~ 7% at 10 years) [62],other recent recently published data indicate that of patientswho had elective surgery for diverticulitis, 25% experiencedpersistent abdominal symptoms [74]. Neither the stage ofdisease (complicated or uncomplicated) nor the surgicaltechnique (laparotomy or laparoscopy) was significantlyrelated to the occurrence of symptoms [75]. A more individu-alized approach taking into account the frequency, severityof the attacks, and their impact on quality of life should guidethe indication for surgery [73].
4.3.4 5-Aminosalycilic acidBecause mesalazine was effective in controlling SUDD and inpreventing diverticulitis occurrence from SUDD [42,43], its usein preventing diverticulitis recurrence has been consideredmore interesting. Seven double-blind, placebo-controlledstudies have recently concluded in assessing the role ofmesalamine in preventing recurrence of diverticulitis [75-80].Unfortunately, most of them did not find mesalazine to besignificantly superior to placebo in preventing diverticulitisrecurrence, even if the DIV/04 trial found a decrease in therelative risk of recurrence after 24 months in the mesalazinegroup than in placebo group (11 vs 28%, respectively;p = 0.48, 95% CI 0.20 -- 1.15) [76]. On the other hand, mesa-lazine was found to be significantly better than placebo inreducing abdominal symptoms following acute diverticulitis(DIVA trial: p = 0.045; DIV/04 trial: p = 0.021) [75,76].
Only a trial conducted in Romania found that mesalazine(514.7 ± 30.5 mg/day) was better than placebo in reducing
the risk of developing diverticulitis (p = 0.044) over a40-month period, as well as the number of diverticulitis flares(p = 0.001) and the need for surgery (p = 0.02). The relativerisk of developing diverticulitis was 2.47 times higher (95%CI 1.38 -- 4.43) in the placebo group compared with themesalazine group [80].
4.3.5 ProbioticsAfter the first study published > 10 years ago [81], anotherstudy investigated the combination of balsalazide, a 5-ASApro-drug, with the high-potency, probiotic mixture VSL#3�
(composed by eight different bacterial strain: L. casei,L. plantarum, L. acidophilus, L. delbrueckii subsp. Bulgaricus,B. longum, B. breve, B. infantis, Streptococcus salivarius subsp.thermophilus) in preventing recurrence of diverticulitis. Thecombination balsalazide/VSL#3 was better than VSL#3 alonein preventing relapse of uncomplicated diverticulitis of thecolon, even if without statistical significance (73.33 vs 60%,p < 0.1) [82].
Finally, an interesting approach in preventing diverticulitismay be the use of bacterial lysate. We know that intestine iswell known as the largest human lymphoepithelial organand daily produces more antibodies, mainly secretory IgA,than do all other lymphoid tissues. Dughera et al. assessedthe efficacy of an oral immunostimulant highly purified, pol-ymicrobial lysate (containing 80 � 109 Escherichia coli strains01, 02, 055 and 0111, and 1 � 109 Proteus vulgaris) in theprevention of recurrent attacks of diverticulitis and in theimprovement of symptoms. Eighty-three consecutive patientssuffering from recurrent symptomatic acute diverticulitis andwith at least two attacks in the previous year were randomlyassigned to receive (group A) an oral polybacterial lysate sus-pension (5 ml two times day for 2 weeks every month) or toa no-treatment clinical follow-up as controls (group B) for a3-month follow-up. Statistical differences of the sums of thescores between group A versus group B were recorded after1 month (p < 0.05) and 3 months (p < 0.01) of treatmentwith the oral polybacterial lysate suspension [83].
4.3.6 Nutritional treatmentA recent double-blind, randomized, placebo-controlled studyassessed the role of butyrate in preventing diverticulitis recur-rence [84]. Butyrate is a short-chain fatty acid previouslyproven to provide symptomatic relief in patients sufferingfrom various colonic diseases, ranging from diarrhea to IBD,and suggestive of preventive role in carcinogenesis ofcolonocytes [85-87]. Seventy-three patients with at least one epi-sode of diverticulitis no earlier than 1 year before the studywere randomly treated with microencapsulated sodium buty-rate 400 mg daily or placebo. After 12 months, clinical symp-toms of diverticulitis occurred in two (6.67%) and seven(31.8%) patients in the active and placebo groups, respec-tively (p = 0.0425) [84]. Although limited by some bias, thisstudy supports the hypothesis that therapy that is able toinfluence colonocyte metabolism, reinforcing colonic mucosal
barrier leading to decreased inflammation of the mucosa, andincreasing cell regeneration rate with healing of the mucosa, isanother interesting option in preventing diverticulitisrecurrence.
5. Conclusion
Although antibiotic-based therapies are yet common clinicalpractice for the treatment of uncomplicated DD, the intro-duction of mesalazine and probiotics has changed the medicalapproach to the disease. However, further studies are neededto find optimal doses and optimal therapeutic regimens.
6. Expert opinion
Looking at the new pathophysiological understatement aboutDD, some further comment is needed.
The new approach based on microscopic/macroscopicinflammatory findings in the colon has been described above.The new pathogenetic theory tends to describe DD as achronic disease similar to IBD. But some questions arise dueto this theory.
How might a chronic mucosal inflammation lead to fullthickness bowel wall alterations? What is the relationshipbetween this inflammation and either the symptoms associ-ated with the various phases of the disease or the risk ofrecurrences? We currently know that the mechanism of muco-sal healing in IBD, represented by the network TNF-a,Syndecan1 (SD1) and beta-Fibroblastic Growth factor(bFGF), is related to the severity of inflammation [88]: if thisnetwork fails, a pro-fibrotic behavior can occur as in Crohn’sdisease (CD) [89]. We currently know that DD has a similarbehavior of CD [90]; thus, it is hypothesized that imbalanceof the framework TNF-a/SD1/bFGF may cause low-gradeinflammation of the full thickness colonic wall. Alternatively,we may suppose that such chronic inflammation is theresponse of the mucosa to the high pressure due to the DD.In this case, it would be the high pressure, rather than anunlikely progression from low- to high-grade mucosal inflam-mation, is responsible for the progression from asymptomaticdiverticulosis to diverticular perforation.
The second question is why rifaximin is ineffective inpreventing diverticulitis recurrence. Of course, it may bedue to varying study quality on this specific outcome. Froma pathophysiological perspective, the ineffectiveness of rifaxi-min in preventing diverticulitis recurrence may be caused bythe short-lasting effect of the reduction of fecal bacterialcounts during oral treatment with rifaximin. The bacterialpopulation of the colon (mostly E. coli, Bacterioides spp. andanaerobic cocci) has been shown to recover within1 -- 2 weeks after the end of treatment [91]. Thus, repeatedoral administration of rifaximin (as per the regimen describedby Lanas et al. [69]) may control the colonic bacterial popula-tion for only 15 -- 20 days, with high colonic bacterial
concentrations for the last 10 days of the month and thereforewith high risk of diverticulitis recurrence.The third question is why mesalazine seems to be effective
in preventing SUDD but not in preventing diverticulitisrecurrence. Low-grade inflammation has been clearly identi-fied in biopsy samples taken from SUDD patients [19-23], aswell as significant inflammation has been identified in acutediverticulitis [24-27]. Thus, why those different results? A sim-pler hypothesis is because there are biases, most of these trialsare affected. Heterogeneity in the population enrolled, hetero-geneity in the type of mesalamine investigated linked to themechanism of discharging through the colon, heterogeneityin end points assessed and detected through the publishedtrial all influence these conflicting results. Further studies areneeded to overcome these limits, for example, enrollingpatients with the same endoscopic and/or radiological findingof the disease [15,92].A more intriguing hypothesis is that SUDD and diverticu-
litis are different diseases, as UC and CD are differentdiseases. In fact, we have mucosal inflammation in SUDD
but a full thickening inflammation in diverticulitis. We canhypothesize that mesalazine may be able to penetrate throughthe mucosa but not to the entire colonic wall: it may thereforebe able to treat and prevent SUDD but not diverticulitisrecurrence. Hence, early treatment with mesalazine may beeffective in preventing diverticulitis occurrence or recurrenceas we can hypothesize looking at the results of our recenttrial [43].
Taking all these hypotheses into consideration, currentpitfalls in the treatment of DD and its complications haveto be addressed by future studies.
Declaration of interest
The author has no relevant affiliations or financial involve-ment with any organization or entity with a financial interestin or financial conflict with the subject matter or materialsdiscussed in the manuscript. This includes employment,consultancies, honoraria, stock ownership or options, experttestimony, grants or patents received or pending, or royalties.
BibliographyPapers of special note have been highlighted as
either of interest (�) or of considerable interest(��) to readers.