Multicentre observational study of the natural history of left-sided acute diverticulitis

BACKGROUND

The natural history of acute diverticulitis (AD) is still unclear. This study investigated the recurrence rate, and the risks of emergency surgery, associated stoma and death following initial medical or surgical treatment of AD.

METHODS

The Italian Study Group on Complicated Diverticulosis conducted a 4-year multicentre retrospective and prospective database analysis of patients admitted to hospital for medical or surgical treatment of AD and then followed for a minimum of 9 years. The persistence of symptoms, recurrent episodes of AD, new hospital admissions, medical or surgical treatment, and their outcome were recorded during follow-up.

RESULTS

Of 1046 patients enrolled at 17 centres, 743 were eligible for the study (407 recruited retrospectively and 336 prospectively); 242 patients (32·6 per cent) underwent emergency surgery at accrual. After a mean follow-up of 10·7 years, rates of recurrence (17·2 versus 5·8 per cent; P < 0·001) and emergency surgery (6·9 versus 1·3 per cent; P = 0·021) were higher for medically treated patients than for those treated surgically. Among patients who had initial medical treatment, age less than 40 years and a history of at least three episodes of AD were associated with an increased risk of AD recurrence. There was no association between any of the investigated parameters and subsequent emergency surgery. The risk of stoma formation was below 1 per cent and disease-related mortality was zero in this group. The disease-related mortality rate was 0·6 per cent among patients who had surgical treatment.

CONCLUSION

Long-term risks of recurrent AD or emergency surgery were limited and colectomy did not fully protect against recurrence.

[One-stage curative treatment of neoplastic colonic obstruction: long-term results, comparison with elective surgery and therapeutic implications]

Over a period of ten years (1980-1989) 528 patients with colon cancer were treated at one institution. One hundred seventy nine (33.9%) were obstructed (O) and underwent emergency surgery, while 349 received elective (E) treatment; of these 363 had one-stage curative treatment. Operative mortality was 10.3% (O) and 3.5% (E) respectively (p < .0.5). Three hundred forty three patients survived surgery and entered follow-up: 96 were O (M:F, 54:42) and 247 E (M:F, 119:128, p = N.S.). Their mean age was 69.5 and 64.4 (p < .001), respectively. Dukes' stage and histological grading were evenly distributed within the two groups, but sites of the primary were not (p < .001). During the follow-up local recurrence occurred in 40 patients (13 O, 27 E, p = N.S.) and metastatic disease in 78 (28 O, 50 E, p < .05, Life Table Analysis) including liver recurrence in 17 O and 30 E (p = .063). Five year crude survival (51%) was significantly worse in obstructed patients. Multivariate analysis showed that Dukes' stage and obstruction were the only prognostic factors of recurrent disease, while survival was affected by the same variables and age over 70. When recurrent disease was introduced in the model survival depended on Dukes' stage, site of the primary and age over 70 and the variable obstruction disappeared as prognostic factor. Right sided tumours showed a better and those at the splenic flexure a worse prognosis. Despite one-stage curative treatment obstruction carries a significantly higher risk of developing metastatic disease, suggesting that obstruction enhances cancer cell dissemination. These patients might benefit from per-operative intra-portal and post-operative systemic adjuvant chemotherapy.