Outcomes after medical and surgical treatment of diverticulitis: A systematic review of the available evidence
Abstract
There is still controversy regarding the appropriate management of diverticulitis of the colon in cases when both surgical and conservative treatment may be an option.
We performed a systematic review of the available evidence regarding the outcomes after medical and surgical treatment of diverticulitis from studies published after 1980 and indexed in the PubMed database. We included original studies that reported comparative data for at least one outcome in medically- and surgically-treated patients with transverse or left colon diverticulitis. The main outcomes of interest were mortality, morbidity, and recurrence of diverticulitis after medical or surgical treatment.
There were 21 studies fulfilling our inclusion criteria out of 1360 initially identified as possibly relevant. More patients were treated conservatively in the included studies compared to emergency surgery (24 862 vs 6504). Emergency surgery was the main option for patients with severe complications of diverticular disease, including peritonitis. In most studies, in-hospital mortality for patients treated surgically was generally higher than that of patients treated medically, whereas there were insufficient comparative data regarding mortality during follow up. However, readmission to the hospital due to diverticular disease during follow up was more common in the group of patients treated conservatively compared to those treated surgically (4358/23 446 [18.6%]vs 22/359 [6.1%]). Conservatively-treated patients, with a first or second episode of diverticulitis, required surgery for recurrent disease during follow up in a maximum of 45% of cases, with larger studies reporting percentages lower than 11%.
It should be emphasized that medical and surgical treatments have not ever been compared in a randomized controlled trial in patients with diverticulitis (without generalized peritonitis that is a surgical emergency). Although medical treatment results in more readmissions due to recurrence, it may be reasonable to avoid surgical therapy in the vast majority of patients with acute diverticulitis. It is unclear what the best treatment option is for younger patients (<50 years), namely whether elective surgery should be considered with the first episode of diverticulitis.
Introduction
Colonic diverticula (diverticulosis) are present in an estimated 20–30% of the general population, with increased prevalence in older adults and in populations of Western and industrialized countries.1–5 The frequency of diverticulosis in patients aged 30–39 years is about 5%, whereas in those over 80 years old it may be as high as 60%.1,4,5 Diverticula of the right colon are more common among Eastern populations, whereas in Western populations the left colon is the anatomical location of diverticula in the great majority of cases.2,6 Diverticulitis, defined as inflammation and/or infection of diverticula, is the most common clinical complication of diverticulosis and it is increasing in incidence in populations that have adopted the Western type of diet that includes low fiber.7 It is estimated that approximately 20% of all patients that develop diverticula will have either diverticulitis or bleeding episodes, which is another common complication of diverticular disease. In addition, a significant proportion of patients with diverticulitis may develop complications, including abscess, obstruction, stricture, fistula, or peritonitis.
There is still controversy regarding the appropriate management of the various stages of the disease and its complications. Several investigators and clinicians believe that the conservative, non-operative treatment of patients with diverticulitis is generally successful in cases without generalized peritonitis. On the other hand, there are many supporters of the invasive operative treatment for both uncomplicated and complicated cases of diverticulitis. Historically, there has been a change of the therapeutic approach in patients with diverticulitis. From a conservative approach based on antimicrobial treatment and bowel rest in the 1970s, there was a movement to a more aggressive therapy based on early surgical intervention in the 1980s and 1990s. Nowadays, both therapeutic modalities, medical and surgical, are commonly used.
There are several unanswered questions regarding the management of patients with diverticulitis, such as whether and when to operate in cases without generalized peritonitis, and what the recurrence rate per year is, especially after medical treatment. Unfortunately, medical and surgical treatments have not ever been compared in a randomized controlled trial in patients with diverticulitis (without generalized peritonitis that is a surgical emergency). Thus, we performed a review of the available published evidence regarding the comparative characteristics of the medical and surgical treatment of patients with transverse or left colon diverticulitis from relatively recent studies published after 1980.
Methods
Literature search
We used the PubMed database (1/1980–1/2006) to identify relevant studies for our review. The search strategy was based on the following key words: diverticular disease, diverticulitis, medical treatment, and surgical treatment.
Study selection
We included original studies that reported comparative data for at least one outcome in medically- and surgically-treated patients with transverse or left colon diverticulitis. We excluded studies that focused on patients with right colon diverticular disease or other colonic diseases apart from diverticulitis. However, studies were included in the review if a minority of the patients enrolled in them had right-sided diverticulitis and these patients were excluded from analysis whenever possible. Also, we excluded studies that focused on laparoscopic surgery for the treatment of diverticulitis. We examined studies published after 1980 in an attempt to account for the changes in the standards of medical and surgical therapy during the last decades.
Data extraction
The data we extracted from the original studies that were included in our review were the date of publication, the setting of the study, the patient population, the number of patients included in each therapeutic group (medical or surgical), and morbidity and mortality during admission and follow up.
Definitions
Medical (conservative) treatment consisted of bowel rest, intravenous fluids, and antibiotics. Percutaneous drainage of pericolic abscess (although an invasive procedure), was allowed to be included as part of the conservative treatment of patients with diverticulitis when data for patients treated only by conservative methods were not presented separately in the studies. All relevant surgical procedures, excluding laparoscopic or percutaneous-only techniques, were considered as surgical therapy. Operations performed during the studied admission were considered emergent, including those that followed an initial conservative treatment that failed, whereas planned operations performed during another hospitalization, after conservative treatment during the initial admission, were considered elective. A death was considered attributable to diverticulitis when it was caused during an acute episode and was related directly to the natural course of the disease (i.e. death due to sepsis, hemorrhage, or other complications), or to the therapy itself (e.g. death due to the toxicity of medications or any intraoperative death), or it was the result of dramatic worsening of an already existing condition (e.g. death due to cardiac ischemia after hemorrhage). On the contrary, death caused by an event that could have coincided with the diverticulitis attack (e.g. a heart attack or a stroke in a patient with free medical history and mild diverticulitis) was considered as all-cause mortality. If the authors of a paper had already characterized the reported mortality as attributable to diverticulitis or as all-cause, we accepted their categorization.
Outcomes
The main outcomes of interest in our review were mortality, morbidity, and recurrence of diverticulitis after the medical or surgical treatment of patients with diverticulitis. Morbidity was defined as the development of postoperative complications (including wound abscess, fistula, small bowel obstruction, and stoma complications) that led to re-operation or as any deterioration in a clinical condition of a conservatively-treated patient necessitating an emergency operation during the studied admission. In addition, the need for elective surgery in the management of diverticulitis was evaluated by comparing outcomes in patients treated only conservatively to those in patients treated conservatively initially while planned for an elective surgery. We did not attempt to analyze the effect that different surgical procedures had in the studied outcomes.
Results
We initially identified 1360 studies by our search in PubMed. A large proportion of the studies were excluded as they were irrelevant. In addition, many studies were excluded either because they were not original, but review studies; because they focused only on medical or surgical therapy without reporting comparative data for these two modes of therapy; or because they focused on diverticulitis of the right colon. Finally, there were 21 studies8–28 matching our inclusion criteria (three studies reporting different outcomes of the same set of patients were analyzed as a single16,20,27), which were further analyzed. In Table 1 we present the study characteristics and patient population of studies that focused on the comparison of medical and surgical treatment of patients with diverticulitis.
| First author (year of publication)/Study design | No. patients clinically evaluated | Age in years (mean value unless otherwise stated) | Type of diverticular disease -complications | Emergency surgical treatment (n of patients) | Medical treatment (n planned for elective operation if any) | Type of diverticular disease in patients treated conservatively | No. patients that completed follow up (surgical + medical treatment) | Duration of follow up (years; (mean value unless otherwise stated) |
|---|---|---|---|---|---|---|---|---|
| Ouriel K (1983)/Retrospective cohort | 92† | 34.3 | Uncomplicated or complicated acute diverticulitis except from hemmorhage | 16 | 76 (9) | Uncomplicated acute diverticulitis | 82 (15 + 67) | Surgical patients 1.7, medical patients 2.25 |
| Ambrosetti (1993)/prospective study | 226 | 64 (median) | Acute diverticulitis of the left colon | 66 | 160 | Acute diverticulitis | 160 (0 + 160) | 2.5 (median) |
| Tudor (1994), Morton (1995), & Farmakis (1994)/Multicentre prospective cohort‡ | 300 | 67.7 | Complicated diverticular disease (104 acute phlegmon, 63 peritonitis, 40 hemorrhage) | 176 | 124 | Mainly acute phlegmon and hemorrhage. Only 4 cases of (purulent) peritonitis. | 120 (77 + 43) | ≥5 |
| Sarin (1994)/ Mixed design (retro + pro) | 162§ | 68 (median) | 86 diverticulitis, 31 peritonitis§, 8 obstruction, 37 hemorrhage | 52 | 110¶ | 73 patients diverticulitis, 37 lower GIT hemorrhage¶ | 144 (43 + 101) | 4 (median) |
| Vignati (1995)/Retrospective cohort | 40 | 42.5 | Uncomplicated or complicated acute diverticulitis | 10 | 30 | NR | 28 (0 + 28) | 5-9 (range) |
| Munson (1996)/Retrospective cohort | 65 | 55 | NR | 33 | 32 | NR | 65 (33 + 32) | 1.9 |
| Cunningham (1997)/ Retrospective cohort | 29 | 31 | Uncomplicated or complicated acute diverticulitis | 14 | 15 (4) | Uncomplicated diverticulitis (those operated electively had phlegmon or abscess) | 29 (14 + 15) | Surgical group 2.8, medical group not specified |
| Maggard (1999)/Retrospective cohort | 39 | 46 | Complicated diverticulitis, 9 peritonitis | 34 | 5 (5) | Not peritonitis | NR | Short-term outcomes examined |
| Reisman†† (1999)/Retrospective cohort | 108 | 64 | 108 left or transverse and 11 to the right colon, 30% complicated | 27 | 81 (10) | Uncomplicated diverticulitis or bleeding | 108 (27 + 81) | 3.3 (median) |
| Netri (2000)/Retrospective cohort | 239 | 63 | Uncomplicated or complicated acute diverticulitis | 42 | 197 (44) | Patients without peritonitis (including those with stenosis or fistulas who were operated electively at a later time) | NA | Short-term outcomes examined |
| Biondo (2002)/Retrospective cohort | 327 | 78% of patients were > 50 years old | First attack of uncomplicated or complicated acute diverticulitis | 101 | 226 | Patients without diffuse peritonitis, septic shock, or pneumoperitoneum | 327 (101 + 226) | >2, maximum 7.5 |
| Cavallaro (2002)/ Retrospective cohort | 77 | 70 (median) | III-IV Hinchey | 5 | 72 (10) | I-II Hinchey | NR | NR |
| Bahadursingh (2003)/Retrospective cohort | 192 | 61 | Acute diverticulitis, abscess, diverticulosis, perforation, stricture, fistula | 73 | 119 | First attack of diverticulitis, diverticulosis, comorbid conditions (high risk), patient refusal for surgery | NA | Short-term outcomes examined |
| Guzzo (2004)/Retrospective cohort | 762 | 76% >50 years old | First attack of uncomplicated or complicated acute diverticulitis | 175 | 587 | Uncomplicated diverticulitis | 762 (175 +587) | 5.2 (median) |
| Ambrosetti (2005)/Prospective cohort | 73 | 68 (median) | 45 pts with mesocolic and 28 with pelvic diverticular abscess | 18 | 55 (2)‡‡ | 38 patients had mesocolic and 17 pelvic abscess | 73 | 3.6 (median) |
| Mueller (2005)/Retrospective cohort | 363 | 64 (median) | 252 uncomplicated diverticulitis, 111 complicated diverticulitis | 111 | 252 | All had non-complicated diverticulitis | First follow up 167 patients Second 85 patients | First follow up after a median of 7.2 years, Second after 13.3 years |
| Greenberg (2005)/Retrospective cohort | 49 | 34.4 | Uncomplicated or complicated acute diverticulitis (80% first episode) | 15 | 34 (5) | NR | 49 (15+34) | 6.9 for surgical and 5.72 for medical patients. All patients were followed for >1 year |
| Broderick-Villa (2005)/Retrospective cohort | 3165 | 50% were <60 years old | Initial episode‡‡ of uncomplicated or complicated acute diverticulitis | 614 | 2 551 (185)§§ | NR | 3127 (NR) | 8.9 |
| Anaya (2005)/Retrospective cohort | 25 058 | 69 | Initial episode‡‡ of uncomplicated or complicated acute diverticulitis | 4922 | 20 136 (574) | NR | 20 136 (0+20 136) | 4.3 (median) |
- † Five patients with diverticular bleeding were not presented here because the authors did not report on outcomes of our interest;
- ‡ first two studies report data regarding the initial admission and short-term outcomes, whereas the third report data of patients followed up until their death or for at least 5 years;
- § § two patients diagnosed postmortem as peritonitis due to diverticulitis were excluded from further analysis;
- ¶ ¶ 13 patients that failed initial conservative treatment were operated on;
- †† †† data are presented for patients with left or transverse colon diverticula when available;
- ‡‡ ‡‡ first episode according to hospitals' databases. A small proportion of patients may have had episodes prior to enrolment to the health services organizations described in the studies;
- §§ §§ Computed tomography-guided percutaneous drainage was performed in some patients in addition to fluid or antibiotic therapy
- GIT, gastrointestinal tract; NA, non-applicable, NR, not reported.
The year of publication of the studies ranged from 1983 to 2005. There were variations in the number of episodes of acute diverticulitis that patients had suffered at the time of inclusion in the studies, except four studies that reported data regarding patients with first admission due to diverticulitis (Table 1).10,12,13,18 There were five studies that evaluated the treatment of diverticulitis in younger (<50 years of age) patients.15,17,19,24,28 More patients were treated conservatively in the included studies compared to emergency surgery (24 862 vs 6504). As shown in Table 1, the majority of those treated conservatively belonged to the group of patients with localized, non-complicated diverticulitis or localized peritonitis (Hinchey score I/II), whereas emergency surgery was a more common treatment in studies evaluating patients with complicated diverticular disease.19,27 The percentage of patients treated conservatively initially while planned for an elective surgery ranged from 2.8% to 27%.10,15 The follow up of the patients in the studies varied considerably, since there were studies evaluating the short-term (<1 years) outcomes of therapy,11,19,23 whereas the rest evaluated the long-term outcomes, with one study reporting data for patients after 13.3 years of follow up.
In Table 2 we present data regarding the outcomes of patients treated with emergency surgical treatment compared to those in patients with medical treatment only. As shown, the in-hospital mortality of patients treated surgically was generally higher than that of patients treated medically within the studies. In five of these studies, this difference was statistically significant.8,11,23,26,27 However, mortality during follow up in the two treatment groups was reported in few studies and no conclusion could be drawn from them. As mentioned before, diverticulitis was more severe and more frequently complicated by abscess, fistula, stenosis, or peritonitis in the group of patients treated by emergency surgery in most studies (Table 1). Mortality was very low in both treatment groups in the studies reporting on diverticulitis in young (<50 years old) patients.15,17,19,24,28 The need for operation (or re-operation) due to complications or failure of treatment seemed to be more common in the group of patients treated conservatively. Specifically, the proportion of patients that failed medical treatment and were subsequently operated on ranged from 0% to 25% in the analyzed studies. The respective values for surgical therapy ranged from 0% to 7% (Table 2).
| First author (Year of publication)/Study design | Attributable in-hospital mortality | All cause in-hospital mortality | Attributable mortality during follow up (or during the whole study) | All cause mortality during follow up (or during the whole study) | Complications† requiring operation (or re-operation) during hospitalization | Readmission due to diverticular disease during follow up | Relapse or recurrence of relevant symptoms during follow up | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ST | MT | ST | MT | ST | MT | ST | MT | ST | MT | ST | MT | ST | MT | |
| No. patients with outcome/no. patients in treatment group (%) | ||||||||||||||
| Ouriel K (1983)/Retrospective cohort | 1/16 (6) | 0/76 (0) | 1/16 (6) | 0/76 (0) | 0/15 (0) | 0/76 (0) | 0/15 (0) | 0/76 (0) | NR | NR | 1/15 (7) | 37/67 (55) | NR | NR |
| Ambrosetti (1993)/prospective study | NR | NR | 2/66 (3) | 0/160 (0) | NR | NR | NR | NR | NR | 0/160 (0) | NR | 27/160 (17%) | NR | NR |
| Tudor (1994), Morton (1995), & Farmakis (1994)/ Multicentre prospective cohort‡ | 19/176 (11) | 2/124 (2) | 28/176 (16) | 6/124 (5) | 1/77 (1)§ | 9/43 (21) | NR | NR | NR | NR | 1/77 (1)§ | 14/43 (33) | 2/77 (3)§ | 37/43 (86) |
| Sarin (1994)/ Mixed design (retro + pro) | 6/52 (12)¶ | 1/110 (1) | 6/52 (12)¶ | 1/110 (1) | NR | NR | NR | NR | NR | 13/123 (11) | 0/43 (0) | 10/101 (10) | NR | NR |
| Vignati (1995)/Retrospective cohort | 0/10 (0) | 0/30 (0) | 0/10 (0) | 0/30 (0) | NR | 0/28 (0) | NR | 0/28 (0) | NR | 4/34 (12) | NR | 11/28 (39) | NR | 12/28 (43) |
| Munson (1996)/Retrospective cohort | 0/33 (0) | 0/32 (0) | 0/33 (0) | 0/32 (0) | 0/33 (0) | 0/32 (0) | 0/33 (0) | 0/32 (0) | NR | NR | NR | NR | 9/33 (27) | 20/32 (63) |
| Cunningham (1997)/ Retrospective cohort | 0/18 (0)†† | 0/11 (0) | 0/18 (0)†† | 0/11 (0) | 0/18 (0)†† | 0/11 (0) | 0/18 (0)†† | 0/11 (0) | NR | NR | 3/18 (17)†† | 4/11 (36) | NR | NR |
| Maggard (1999)/Retrospective cohort | 0/36 (0) | 0/5 (0) | 0/36 (0) | 0/5 (0) | 0/36 (0) | 0/5 (0) | 0/36 (0) | 0/5 (0) | 0/36 (0) | 0/5 (0) | 1/36 (3) | 0/5 (0) | NR | NR |
| Reisman†† (1999)/Retrospective cohort | 0/27 (0) | 0/81 (0) | NR | NR | NR | NR | NR | NR | 2/27 (7) | 27/108 (25) | 13/27 (48) | 24/81 (30) | 22/27 (81) | 38/81 (50) |
| Netri (2000)/Retrospective cohort | 3/42 (7) | 0/197 (0) | 5/42 (12) | 0/197 (0) | NR | NR | NR | NR | 3/42 (7) | 22/219 (10) | 3/42 (7)‡‡ | NR‡‡ | NR | NR |
| Biondo (2002)/Retrospective cohort | NR | NR | NR | NR | NR | NR | NR | NR | NR | 23/249 (9) | 0/101 (0) | 52/226 (26) | NR | NR |
| Cavallaro (2002)/ Retrospective cohort | 0/5 | 0/72 | 0/5 | 0/72 | NR | NR | NR | NR | NR | 7/72 (10) | NR | NR | NR | NR |
| Bahadursingh (2003)/ Retrospective cohort | 3/73 (4) | 0/119 (0) | 4/73 (6) | 0/119 (0) | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR |
| Guzzo (2004)/Retrospective cohort | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR |
| Ambrosetti (2005)/ Prospective cohort | NR | NR | NR | NR | (2/18 [11]) | (0/55 [0]) | (8/18 [44]) | (15/55 [27]) | NR | 18/73 (25) | NR | 23/55 (42) | NR | NR |
| Mueller (2005)/ Retrospective cohort | NR | 0/252 (0) | 7/111 (6) | 0/252 (0) | NR | 1/252 (0) | NR | 85/252 (34) | NR | 0/252 (0) | NR | 16/167 (10)§§ | NR | 78/167 (47)§§ |
| Greenberg (2005)/Retrospective cohort | 0/15 (0) | 0/34 (0) | 0/15 (0) | 0/34 (0) | 0/15 (0) | 0/34 (0) | 0/15 (0) | 0/34 (0) | NR | NR | NR | NR | 3/20 (15) | 16/29‡‡ (55) |
| Broderick–Villa (2005)/Retrospective cohort | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | NR | 314/2 366 (13) | NR | NR |
| Anaya (2005)/Retrospective cohort | NR | NR | 153/4922 (3)§§ | 544/20 136 (3)§§ | NR | NR | NR | NR | NR | NR | NR | 3 826/20 136 (19) | NR | NR |
- † Includes patients that failed initial medical treatment and were subsequently operated due to persistence of symptoms and deterioration. If such patients required re-operation due to further complications, this would be also counted in the group of complications of the surgically-treated patients;
- ‡ first two studies report data regarding the initial admission and short-term outcomes whereas the third report data of patients followed up until their death or for at least 5 years;
- § § authors compare resection (considered as surgical treatment) with medical therapy or surgery other than resection (e.g. diagnostic laparotomy or drainage);
- ¶ ¶ two patients diagnosed postmortem as peritonitis due to diverticulitis were excluded from further analysis;
- †† †† authors examine patients that underwent emergency or elective surgery;
- ‡‡ ‡‡ surgical patients were admitted due to late surgical complications. Authors reported only on readmissions for elective surgery in the medical therapy group;
- §§ §§ refers to the first follow up of the study.
- MT, medical therapy; NR, not reported; ST, surgical therapy.
Readmission to the hospital due to diverticular disease during follow up was found to be more common in the group of patients treated conservatively compared to those treated by emergency surgery (4358/23 446 [18.6%]vs 22/359 [6.1%]). When we excluded the two larger studies of the analysis,10,13 since they did not report data for the surgically-treated patients for this outcome, the result remained the same (218/944 [23%]vs 22/359 [6.1%]). Similarly, this result remained unchanged when only studies reporting the readmission in both treatment groups were analyzed (141/534 [26.4%]vs 19/317 [6%]). The proportion of patients readmitted ranged from 0% to 55% and from 0% to 48% for the medical- and surgical-treatment groups, respectively. The recurrence of symptoms compatible with diverticular disease not leading to readmission during follow up was reported in relatively few studies as shown in Table 2. Recurrence ranged from 43% to 86% and from 3% to 81% in the medical and surgical groups, respectively.
In Table 3 we present data from studies that reported information regarding patient populations and outcomes after conservative treatment with or without a succeeding elective operation. Mortality during treatment and follow up was zero in both treatment arms in two studies that examined this outcome.17,24 However, the studies reported some complications in the electively-treated patients.17 As shown, only conservatively-treated patients with a first or second episode of diverticulitis required surgery during follow up in a maximum of 45% of cases, with larger studies reporting percentages lower than 11%.10,11 The proportion of necessary emergency operations in these patients varied from 3% to 45% among the studies, with the largest study reporting that 4% of the patients had an emergency operation during follow up.10
| First author (year of publication) | Total no. patients in study | Age in years (mean value unless otherwise stated) | Duration of follow up (years) (mean value unless otherwise stated) | No. episodes of diverticulitis | No. patients planned for elective surgery/total no. patients treated conservatively (%) | Proportion of patients only treated conservatively who underwent surgery during follow up (%) | Proportion of patients only treated conservatively who underwent emergency surgery during follow up (%) |
|---|---|---|---|---|---|---|---|
| Ouriel 1983 | 115 | 34.3 | 0.8 for surgical and 2.25 for medical | NR | 9/76 (12) | 30/67 (45) | 16/67 (24) |
| Biondo 2002 | 327 | 78% of patients were >50 years old | >2, maximum 7.5 | Second episode | 33/44 (75)† | 3/11 (27) | 2/11 (18) |
| Guzzo 2003 | 259‡ | Patients <50 years | 5.2 (median) | First episode | 36/196 (18) | 5/160 (3) | 5/160 (3) |
| Broderick-Villa 2005 | 3165 | 50% of patients were <60 years old | 8.9 | First episode§ | 185/2551 (7) | <222/2366 (9)¶ | NR |
| Greenberg 2005 | 49 | 34.4 | 6.9 for surgical and 5.72 for medical patients | First episode | 5/34 (15) | 11/29 (38) | 4/29 (14) |
| Anaya 2005 | 25 058 | 69 | 4.3 (median) | First episode§ | 574/20 136 (3) | 2202/19 562 (11) | 692/19 562 (4) |
- † Four patients that were discouraged from elective operation due to increased age and high risk were excluded from our analysis;
- ‡ out of a total 762 patients 259 were aged <50 years and for those patient data were available for analysis;
- § § first episode according to hospitals' databases. A small proportion of patients may have had episodes prior to enrollment to the health services organizations described in the studies;
- ¶ 314/2366 (13%) conservatively-treated patients had a readmission during follow up. Authors report that the proportion of them that had re-recurrence (92 patients) were all treated conservatively, but do not provide data for those that had only one recurrence (222 patients).
- NR, not reported.
Discussion
The available data from the studies included in our review support that current medical and surgical therapy of acute diverticulitis is effective in the great proportion of patients and that mortality is generally low in both treatment groups. In addition, reported mortality in both treatment groups was almost zero in the studies that evaluated younger patients (<50 years).15,17,19,24,28 Although there are no definitive criteria regarding the choice of therapy, the type of disease (complicated vs uncomplicated), the age of the patient, and his/her general medical condition are the main factors that influence the physicians' choice. Patients with complicated forms of diverticular disease and especially peritonitis were treated by surgery in almost all of the included studies. In addition, in some studies, it was clarified that patients with a bad general medical condition were preferentially treated conservatively.8,20 In studies that evaluated patients of younger age, the proportion of patients treated surgically (or conservatively while planned for an elective surgery) was generally high.15,17,19,24,28
The analysis of the included studies showed that recurrence and especially readmission due to diverticular disease was more common in patients treated conservatively compared to those treated surgically within the studies. Parks et al., in their cornerstone study regarding the natural course of diverticular disease, reported that readmission due to recurrent diverticulitis was 25% in 317 conservatively-treated patients.29 The respective proportion in our analysis was 18.6%, showing that current medical therapy has not substantially decreased the recurrence of diverticular disease. However, given the fact that the prevalence of symptomatic diverticular disease is increasing in the Western world and especially among younger patients, one may postulate that the proportion of recurrence in our analysis should have been higher than that of the older series. Thus, although recurrence has only slightly decreased, this may actually represent a success for the current medical treatment.
Nevertheless, one may claim that surgery is probably the preferable management strategy for patients with diverticulitis who have risk factors for frequent recurrences of the disease. On the basis of the assumption that the rate of recurrence after an episode of diverticulitis is similar between the various age groups, it is expected that younger people, who have a longer life expectancy, will also have more chance of developing recurrences of diverticulitis during their lives. Earlier studies have confirmed this increase in the recurrence of acute diverticulitis and have also reported that diverticulitis in young patients may have a more aggressive course.29–31 Many studies have been performed to examine the above hypotheses in young patients and the results seem rather contradictory, especially regarding the virulence of diverticular disease in younger patients.25,28,32
Our review was not designed to examine the aforementioned questions specifically in young patients with diverticulitis. Nevertheless, in the studies that followed up young patients with diverticulitis for a long period of time, the reported readmissions due to recurrence were higher than the average proportion of recurrences from all studies, both for surgical15,24 and medical therapy.15,24,28 In addition, Ambrosetti et al.,8 Anaya et al.,10 and Broderick Villa et al.13 found that the relapse of acute diverticulitis or the development of complications after discharge in the medical group was statistically and significantly more common in younger patients, whereas Biondo et al.,12 in a study designed to examine this question, found no difference. It is noteworthy that the proportion of conservatively-treated young patients that were ultimately operated on during follow up varied considerably in the three studies included in our review that reported on this outcome.17,18,24
Data from the large study by Broderick-Villa et al.13 support that that the recurrence rate after an initial episode of diverticulitis treated medically is about 1.5% per year. Also, the mean age of patients with the first episode of diverticulitis is approximately 65 years, and such patients have an average life expectancy of 14 years. Thus, the calculated probability for a typical patient with the first episode of diverticulitis to develop recurrence of the disease during his/her rest of the life is approximately 21%. These facts suggest that if an operation had been performed for all patients with the first episode of diverticulitis, then the surgical management of diverticulitis would not have offered any specific advantages compared to medical treatment in 79% of these patients. However, as mentioned before, this is not true for younger patients in whom life expectancy, and subsequently, the calculated probability to develop recurrence of the disease is higher. In addition, Anaya et al.10 found that even after adjusting for age, younger patients (<50 years) are at a 46% higher risk (hazard ratio 1.46) for recurrent hospitalizations compared to older patients.10
All of the above findings must be evaluated in the context of the benefits that may be achieved by a conservative approach, even for young patients. An overview of the analyzed studies shows that in the majority of patients presenting with a form of diverticulitis for which medical therapy is an option (i.e. uncomplicated acute diverticulitis or complications, such as hemorrhage and small abscesses), surgery may be avoided. Data from the studies focusing on young patients suggested that less than 45% of the patients treated only conservatively when elective treatment was an option would eventually need operative treatment during follow up.24 In addition, less than 24% would need to undergo emergency surgery for diverticular disease.24 Furthermore, it should be noted that although mortality in patients treated by elective surgery is generally very low, the studies report some complications in this group of patients.17 Finally, the possible need for a stoma after surgery, which is a feared situation especially in the younger age groups, is gradually diminishing since resection and primary anastomosis during the same operation is becoming more widely used.19
Our study has several limitations. The most important limitation is that the comparative studies we included in our review were not randomized controlled trials, and thus a safe conclusion regarding the comparative success of the conservative and surgical management of patients with diverticulitis cannot be made. Also, the available data allowed for comparisons between patients treated surgically on an emergency basis and those treated medically, but there were not enough data regarding the comparison of the outcomes of patients undergoing elective surgery and those receiving medical treatment. Patients treated surgically on an emergency basis are usually more ill (since a significant proportion of them suffer from generalized peritonitis) compared to patients undergoing medical therapy. Thus, especially regarding the comparison of mortality, the analysis of data serves mainly descriptive purposes and does not intend to provide a direct comparison between the effectiveness of surgical and medical treatments. Another limitation of our review is that the populations in the included studies presented considerable heterogeneity regarding age and localization and the severity of the disease. In addition, there were no homogeneous predefined criteria for the type of treatment provided to the patients in the included studies, with the exception of generalized peritonitis that is by definition a surgical emergency. Furthermore, the follow up of patients varied considerably between the studies. Finally, the proportion of patients treated by different surgical techniques varied between the studies.
Conclusively, current medical and surgical therapy of acute diverticulitis is effective for the great proportion of patients, and mortality is generally low. Although mortality seems to be higher in patients undergoing emergency surgery than those treated conservatively, this finding should be attributed to the larger proportion of patients with severe forms of disease in the first group. Medical treatment results in more readmissions due to the recurrence of disease. However, it is reasonable to avoid surgical therapy in the vast majority of patients with acute diverticulitis and its milder forms of complications. This is because the calculated probability for a typical patient (older than 60 years) with the first episode of diverticulitis to develop recurrence of the disease during his/her rest of the life is low. It is unclear what the best treatment option for younger patients (<50 years) is and whether elective surgery after the first episode should be considered.
