Researchers at Fukuoka University Hospital analyzed data from 320 patients who underwent low anterior resection for rectal tumors between 2016 and 2025. Their findings, published in Annals of Gastroenterological Surgery, used a statistical technique called exploratory factor analysis to dissect the two most widely used patient-reported measures of postoperative bowel function: the LARS score and the Cleveland Clinic Florida Fecal Incontinence Score, better known as the Wexner score. What emerged was a three-part structure that the authors say should change how clinicians think about, measure and treat bowel dysfunction after rectal surgery.

The logic behind the study is straightforward but powerful. The LARS score bundles five symptoms into a single number, while the Wexner score focuses mainly on incontinence. Clinicians have long noticed that the two instruments sometimes disagree, giving conflicting impressions of the same patient. The Japanese team suspected that this discordance arises because the scores are actually measuring different underlying dimensions of bowel dysfunction, dimensions that a composite total quietly blends together. To test that idea, they went beneath the total scores and examined the individual symptom items themselves.

First, the researchers mapped how the symptoms correlated with one another. Some pairs were tightly linked: Wexner liquid incontinence and LARS liquid stool incontinence correlated strongly, with a Spearman coefficient of 0.71, and urgency tracked closely with clustering at 0.59. Other symptoms, notably bowel frequency, floated loosely, correlating weakly with nearly everything else. When the team reorganized the correlation matrix using hierarchical clustering, related symptoms visibly grouped together, hinting that the syndrome had natural fault lines running through it.

Formal factor analysis confirmed the hunch. Using maximum likelihood extraction with Quartimin rotation on pooled observations from 6, 12 and 24 months after surgery, the analysis retained three factors that together explained 64.7 percent of the total variance. The first factor gathered the incontinence-related items, including liquid and solid stool incontinence, pad use and lifestyle alteration. The second was defined almost entirely by urgency and clustering, with a smaller contribution from bowel frequency. The third captured gas incontinence from both instruments, pointing to a separate domain of impaired gas control that neither score was designed to isolate on its own.

The real payoff came when the team asked which clinical factors drove each domain. Using multivariable mixed-effects models that accounted for repeated measurements in the same patients over time, they found that the same risk factors did not affect all domains equally. Very low anterior resection, in which the stapled anastomosis is created inside the anal canal itself, and a transanal surgical approach strongly worsened the incontinence domain, with beta coefficients of 0.437 and 0.860 respectively, but had much weaker effects on urgency and clustering. This pattern fits the anatomy: an ultra-low anastomosis and transanal dissection can directly injure the sphincter complex, undermining the mechanical barrier against leakage rather than the reflexes governing urgency.

Radiotherapy told a different story. Preoperative radiation worsened all three domains, suggesting it damages pelvic tissue in a broader, less selective way. Preoperative chemotherapy raised incontinence severity as well. Meanwhile, the age findings were the most striking and perhaps the most surprising. Compared with patients under 50, those aged 50 to 70 had greater incontinence-domain severity, consistent with an age-related decline in pelvic floor reserve. Yet patients aged 70 and older actually reported less urgency and clustering than the youngest group, with a negative beta coefficient of -0.257. The authors interpret this inversion as evidence that the two domains arise from different biology: incontinence from structural, anatomical vulnerability, and urgency with clustering from functional abnormalities, such as heightened visceral sensitivity and colonic hypermotility, that resemble diarrhea-predominant irritable bowel syndrome and tend to fade with age.

That distinction has immediate therapeutic implications. Patients whose suffering is dominated by urgency and clustering, often younger people, may respond to drugs that calm colonic hypermotility; the authors point to emerging evidence that 5-HT3 receptor antagonists such as ramosetron can relieve urgency and frequent stools in selected LARS patients. Patients with incontinence-dominant symptoms, by contrast, are more likely to benefit from physical rehabilitation of the pelvic floor and sphincter, including biofeedback and pelvic floor muscle training. A single composite score, the study argues, cannot tell these patients apart, because two people with identical LARS totals may occupy entirely different symptom worlds.

The study has important caveats, which the authors lay out candidly. It was conducted at a single center, and the three-factor structure was derived exploratively without confirmation in an independent cohort, so the domains remain hypothesis-generating rather than established. The LARS items were reweighted onto an ordinal severity scale before analysis, since the original score’s non-linear item weights were built for classification, not structural modeling. Questionnaire response fell to 66.9 percent at 24 months, and missing data were handled under a missing-at-random assumption without formal sensitivity analysis, meaning late symptom severity may have been underestimated. The analysis also excluded obstructed defecation and quality-of-life measures, which international consensus definitions consider integral to the full LARS construct, and the classification of very low anterior resection relied on the operating surgeon’s intraoperative judgment rather than a standardized anatomical measurement.

Even with those limitations, the message is hard to ignore. Postoperative bowel dysfunction, long summarized by a single number on a questionnaire, appears to be a family of related but separable conditions, each with its own anatomy, epidemiology and treatment logic. A domain-based approach, the authors conclude, offers a more precise framework for individualized functional assessment and could guide the design of future intervention trials, whether testing antihypertensive-style motility drugs for young patients with urgency-dominant disease or targeted pelvic floor rehabilitation for those with structural incontinence. For rectal cancer survivors, the difference between one blurred score and three clear domains may ultimately be the difference between generic reassurance and treatment that actually fits the problem.

Subject of Research: Symptom domain structure of postoperative bowel dysfunction after sphincter-preserving rectal cancer surgery

Article Title: Exploring Symptom Domains of Postoperative Bowel Dysfunction by Integrating LARS and Wexner Scores: A Longitudinal Factor‐Analytic Study

Article References: Matsumoto, Y., Takeshita, I., Shiokawa, K., Sahara, K., Munechika, T., Nagata, K., Nagano, H., Takahashi, H., & Hasegawa, S. (2026). Exploring Symptom Domains of Postoperative Bowel Dysfunction by Integrating LARS and Wexner Scores: A Longitudinal Factor‐Analytic Study. Annals of Gastroenterological Surgery, Article ags3.70289. https://doi.org/10.1002/ags3.70289

Image Credits: AI Generated

DOI: 10.1002/ags3.70289

Keywords: low anterior resection syndrome, LARS score, Wexner score, rectal cancer surgery, fecal incontinence, factor analysis, bowel dysfunction, urgency, pelvic floor, radiotherapy, patient-reported outcomes, mixed-effects models