Outcomes of Gallbladder Polyps and Their Association With Gallbladder Cancer in a 20-Year Cohort

Abstract

Importance: Gallbladder polyps (GP) are found in more than 4% of adult abdominal ultrasonographs. Their growth pattern and association with gallbladder cancer (GBC) are poorly defined.

Objective: To determine the growth pattern of GPs and their association with GBC.

Design, setting, and participants: This cohort study included 622 227 adult members (ie, aged 18 years or older) of Kaiser Permanente Northern California, an integrated health care delivery system, enrolled between January 1, 1995, and December 31, 2014. The GBC cohort comprised a total of 365 adults with GBC and prior ultrasonography, and the GP cohort comprised 35 970 adults with GPs present on ultrasonography. Data analysis was performed from March 2016 to November 2019.

Exposures: Gallbladder polyps (quantitative size, <6 mm, 6 to <10 mm, and ≥10 mm or qualitative size [ie, tiny, small, moderate, and large]).

Main outcomes and measures: For the GBC cohort, proportion of patients with GBC with polyps identified on preceding ultrasonograph. For the GP cohort, rates of GBC among those with polyps according to size and rate of GP growth of at least 2 mm over time.

Results: The GBC cohort comprised 365 individuals (267 [73.1%] women; 173 [47.4%] white patients; median [interquartile range] age, 71 [61-79] years). After excluding 14 patients who did not have evaluation of polyp size, the final GP cohort comprised 35 856 adults, with 18 645 (52.0%) women, a median (interquartile range) age 50 (40-60) years, and 15 573 (43.3%) white patients. Gallbladder polyps were found in 22 patients (6.0%) in the GBC cohort and in 35 870 of 622 227 adults (5.8%) who underwent abdominal ultrasonography. Of these, 19 (0.053%) were diagnosed with GBC, similar to those without GP (316 of 586 357 [0.054%]). The unadjusted GBC rate per 100 000 person-years was 11.3 (95% CI, 6.2-16.3) overall and increased with polyp size, from 1.3 (95% CI, 0-4.0) with initial size of less than 6 mm (n = 17 531) to 128.2 (95% CI, 39.4-217.0) with initial size of 10 mm or larger (n = 2055). In those observed for at least 1 year, the rate was 3.6 (95% CI, 0.7-6.5) per 100 000 person-years. In 6359 patients with evaluable follow-up, unadjusted cumulative probabilities of polyp growth of at least 2 mm at 10 years were 66.2% (95% CI, 62.3%-70.0%) in polyps initially less than 6 mm and 52.9% (95% CI, 47.1%-59.0%) in polyps initially 6 mm to less than 10 mm.

Conclusions and relevance: In this study, GBC rates were low and similar among patients with and without GPs. Growth of 2 mm or more appeared to be part of GP natural history. The results call into question the strategy of proactively following GP to detect GBC.

Figure 1.. Study Flow Chart

Figure 2.. Cumulative Probability of Polyp Growth

Kaplan-Meier curves show unadjusted risk of growth of existing gallbladder polyps in adults with quantitative initial polyp size. Follow-up was until the last abdominal ultrasonograph. Probability of growth of at least 2 mm in all adults (A; log-rank tests results for overall comparison: P < .001; <6 mm vs 6 to <10 mm: P = .01; <6 mm vs ≥10 mm: P < .001), after stability for 3 years (B; long-rank test results for overall comparison: P = .004; <6 mm vs ≥10 mm: P = .03), and after stability for 5 years (C; log-rank tests results for overall comparison: P = .03; <6 mm vs ≥10 mm: P = .03). Probability of polyp size reaching 10mm in all adults with quantitative initial polyp size of less than 10 mm (D; log-rank P < .001), after stability for 3 years (E; log-rank P < .001), and after stability for 5 years (F; log-rank P = .001).