Search

Objective: Patients with functional constipation can be classified according to symptoms and physiological parameters as either having a disorder of defecation or having normal defecation. It is hypothesized that the disordered defecation, where it exists, is a causative factor of the constipation. However, the utility of this classification has yet to be proven in terms of predicting response to therapy. The definitions are non-specific and based on tests that are done in an artificial setting and with derived normal ranges. It is therefore possible that the symptoms and physiological parameters of a defecatory disorder may occur as a continuous spectrum in these patients, rather than defining a discrete entity or subtype. The aim of this study was to use cluster analysis and factor analysis of defecatory symptoms and physiological parameters to look for evidence of subgroups in patients with functional constipation., Material and Methods: Consecutive patients presenting to a specialist constipation clinic and satisfying the inclusion criteria were assessed to determine the severity of defecatory symptoms, and underwent isotope defecating proctography and the Sitzmark transit study. Assessments were made contemporaneously and results of any test not performed within 6 weeks of the initial assessment were excluded. Principle components analysis and cluster analysis were performed to look for evidence of subgroups. Relationships between evacuatory symptoms, index parameters, and test results were explored. The detailed and unselected nature of the analyses produced hundreds of test results, and statistically significant results were critically evaluated in this context., Results: A total of 116 patients were studied (age range 18-73 years, mean 40.5 years). Based on the results of the transit study and proctography, 38% of patients showed evidence of slow transit constipation, 20% FDD (functional defecation disorder), 29% both, and 12% neither. Principle components analysis did not demonstrate an obvious dimension reduction for the variables tested. Cluster analysis (over 150 solutions tested) failed to show evidence of clustering. There were no useful predictive relationships between evacuatory symptoms, index parameters and test results., Conclusions: We used multiple statistical analyses to look for clustering and predictive relationships between clinical and physiological parameters in consecutive patients with functional constipation and found no evidence of the existence of a subgroup of patients with a defecatory disorder. This may be due to weaknesses in the study design, poor validity of the assessments performed, or that defecatory features do not identify a distinct pathophysiological entity, but rather are manifested variably as a continuous spectrum.