Tertiary peritonitis (recurrent diffuse or localized disease) is not an independent predictor of mortality in surgical patients with intraabdominal infection.

BACKGROUND: It is well documented that tertiary peritonitis is associated with different microbiological flora and worse outcomes than secondary peritonitis. It is unknown, however, if these differences can be explained simply by the nosocomial nature of tertiary peritonitis and underlying severity of illness.
METHODS: We reviewed all episodes of intraabdominal infection on the inpatient surgical services at a university hospital over a 46-month period. Univariate analysis and logistic regression were used to compare 91 episodes of secondary peritonitis that progressed to tertiary peritonitis (recurrent diffuse or localized intraabdominal infection) to all episodes of secondary peritonitis (n = 453) to identify predictors for developing tertiary peritonitis. Logistic regression was also used to identify predictors of mortality among patients with secondary (n = 473) or tertiary peritonitis (n = 129).
RESULTS: Of 602 episodes of intraabdominal infection identified, there were 473 episodes of secondary peritonitis, including 20 patients who died within seven days of diagnosis. A total of 129 episodes of tertiary peritonitis were identified, of which 91 were preceded by a single episode of secondary peritonitis, and 38 were preceded by an episode of secondary peritonitis and at least one prior episode of tertiary peritonitis. Tertiary peritonitis was associated with a high APACHE II score (14.9 +/- 0.7), pancreatic or small bowel source, drainage only at initial intervention, gram-positive and fungal pathogens, and a high mortality rate (19%). Increasing APACHE II score (OR 1.07, 95% CI 1.03-1.16, p = 0.0009) independently predicted progression from secondary to tertiary peritonitis while increasing age (OR 0.98, 95% CI 0.97-0.99, p = 0.01) and appendiceal source (OR 0.12, 95% CI 0.02-0.68, p = 0.02) predicted non-progression to tertiary peritonitis. Independent predictors of mortality in this population included increasing age (OR 1.06, 95% CI 1.03-1.1, p < 0.001), increasing APACHE II score (OR 1.18, 95% CI 1.11-1.3, p < 0.001), and four comorbidities: cerebrovascular disease (OR 4.3, 95% CI 1.4-13.1, p = 0.01), malignant disease (OR 2.9, 95% CI 1.3-6.5, p = 0.01), hemodialysis dependency (OR 3.8, 95% CI 1.3-11.2, p = 0.02), and liver disease (OR 4.2, 95% CI 1.6-15.1, p = 0.03). Tertiary peritonitis was not an independent predictor of mortality.
CONCLUSIONS: We were unable to demonstrate, when compared to secondary peritonitis, that tertiary peritonitis is a significant independent predictor of mortality when other variables are taken into account. This suggests that the high mortality associated with tertiary peritonitis is more a function of the patient population in which it occurs than the severity of the pathologic process itself.