P0864 - Association Between Polycystic Ovarian Syndrome and Non-Alcoholic Fatty Liver Disease Fibrosis Severity: A Retrospective Cohort Analysis

Sunday, October 24, 2021

3:30 PM – 7:00 PM PT

Location: Shoreline Exhibit Hall

Has Audio

Award: ACG/Naomi Nakao Gender-Based Research Award
Award: Presidential Poster Award

Presenting Author(s)

Sindhura Kolli, MD

New York University School of Medicine
New York, NY, United States

Award: ACG/Naomi Nakao Gender-Based Research Award

Award: Presidential Poster Award

Sindhura Kolli, MD¹, Sharnendra Sidhu, MD¹, Soonwook Hong, MD¹, Kristen Dammeyer, ¹, Eli Rapoport, ¹, Vlad Obsekov, MD¹, Gabrielle Maranga, MPH¹, Holly F. Lofton, MD¹, Violeta Popov, MD, PhD²
¹New York University School of Medicine, New York, NY; ²NYU Langone School of Medicine, New York, NY

Introduction: Non-alcoholic fatty liver disease (NAFLD) is present in 36-70% of women with polycystic ovary syndrome (PCOS). Both are highly prevalent in subjects with obesity. Androgen overproduction in PCOS promotes a pro-apoptotic environment and may contribute to NAFLD. Our objective was to examine the association between PCOS and NAFLD diagnosis and if PCOS is an independent predictor of advanced fibrosis in patients with NAFLD.

Methods: In a single-center retrospective analysis of electronic medical records, 625 adult patients ( >18 years old) with a diagnosis of NAFLD from 2018-2019 were divided into 3 cohorts: a female study group with PCOS, female control group without PCOS, and a male control cohort, age-matched to the study group. PCOS diagnosis was based on established PCOS society criteria. Our primary outcome was to assess the stage of liver fibrosis, as defined by histology, Fibroscan, MR elastography, NAFLD fibrosis score (NFS), by the age of initial diagnosis. Additionally, demographic, laboratory and clinical parameters were analyzed to compare the three cohorts. Demographics were analyzed using ANOVA, Pearson’s chi-squared, and Kruskal-Wallis methods. Linear regression modeling NAFLD score as a function of age of diagnosis was performed.

Results: A total of 625 subjects with NAFLD were seen. Of these, 21 met criteria for the female NAFLD/PCOS study group, 525 in the female NAFLD only control group, and 79 age-matched male subjects with NAFLD. The NAFLD/PCOS females study group were significantly younger than the other two cohorts at time of diagnosis of NAFLD, had the highest BMI of 38.4, highest AST/ALT ratio of 0.92, lowest albumin value of 4.03, highest percentage of patients with physical signs of cirrhosis or positive signs on CT or MRI of the 3 groups, and a worse NFS than the male cohort. Though not significant, they had the highest Fibroscan steatosis score of 330.50 dB of the three cohorts and a higher fibrosis score of 7.22 kPa than the male group (Table 1). In a linear regression, the male cohort was associated with a 0.7 reduction in NFS compared to women with NAFLD only. Females with NAFLD demonstrate significant difference in the mean NFS and mean age of diagnosis compared to the other 2 groups, while females with both versus males with NAFLD did not (Image 1).

Discussion: Females with PCOS significantly demonstrated NAFLD at an earlier age supported by positive physical and radiological signs, and worse NFS vs. males at the time of initial NAFLD diagnosis.

Figure: Image 1: Subgroup Analyses of Patients with NAFLD by (A) NAFLD Score and (B) Age of NAFLD Diagnosis¹

¹The Female without PCOS subgroup were significantly older and had more comorbidities than the Female with PCOS and Male subgroups

	Total (N=625)	NAFLD Females (N=525)	NAFLD + PCOS Females (N=21)	NAFLD Males (N=79)	p-value
Age	56.93 ± 14.51	60.55 ± 12.41	36.33 ± 9.91	38.37 ± 8.54	< 0.001***
Gender					N/A
Female	546 (87.4%)	525 (100.0%)	21 (100.0%)	0 (0.0%)
Male	79 (12.6%)	0 (0.0%)	0 (0.0%)	79 (100.0%)
Race					0.006
Caucasian	364 (58.2%)	311 (59.2%)	9 (42.9%)	44 (55.7%)
Black	38 (6.1%)	29 (5.5%)	4 (19.0%)	5 (6.3%)
Asian	58 (9.3%)	51 (9.7%)	2 (9.5%)	5 (6.3%)
Hispanic	117 (18.7%)	101 (19.2%)	5 (23.8%)	11 (13.9%)
Hawaiian or Pacific Islander	6 (1.0%)	5 (1.0%)	0 (0.0%)	1 (1.3%)
Native American	4 (0.6%)	4 (0.8%)	0 (0.0%)	0 (0.0%)
Other	18 (2.9%)	9 (1.7%)	1 (4.8%)	8 (10.1%)
Unknown	20 (3.2%)	15 (2.9%)	0 (0.0%)	5 (6.3%)
Age of NAFLD Diagnosis	48.83 ±14.39	53.48 ± 12.19	31.05 ± 8.86	33.49 ± 9.02	< 0.001***
BMI (kg/m2)	32.77 ± 7.08	32.35 ± 6.99	38.40 ± 6.74	34.67 ± 6.91	< 0.001***
Physical Exam Findings of Cirrhosis					< 0.001***
Yes	56 (9.0%)	49 (9.3%)	2 (9.5%)	5 (6.3%)
No	71 (11.4%)	0 (0.0%)	0 (0.0%)	71 (89.9%)
Missing	498 (79.7%)	476 (90.7%)	19 (90.5%)	3 (3.8%)
Platelets	240.11 ± 81.03	237.25 ± 82.15	287.05 ± 69.83	245.99 ± 72.80
AST	52.02 ± 50.64	51.84 ± 52.78	49.49 ± 41.38	53.82 ± 37.51
ALT	65.63 ± 65.80	61.16 ± 64.07	70.05 ± 68.96	92.87 ± 69.99	< 0.001***
AST/ALT Ratio	0.93 ± 0.50	0.97 ± 0.50	0.92 ± 0.48	0.65 ± 0.35	< 0.001***
Albumin	4.20 ± 0.56	4.15 ± 0.54	4.03 ± 0.93	4.54 ± 0.42	< 0.001***
HgbA1c	6.44 ± 1.58	6.55 ± 1.62	6.06 ± 1.12	5.91 ± 1.28	0.008**
US Findings of NAFLD					0.73
Yes	369 (59.0%)	296 (56.4%)	13 (61.9%)	60 (75.9%)
No	16 (2.6%)	13 (2.5%)	0 (0.0%)	3 (3.8%)
Missing	240 (38.4%)	216 (41.1%)	8 (38.1%)	16 (20.3%)
CT Findings of NAFLD					0.024*
Yes	152 (24.3%)	133 (25.3%)	6 (28.6%)	13 (16.5%)
No	17 (2.7%)	12 (2.3%)	0 (0.0%)	5 (6.3%)
Missing	456 (73.0%)	380 (72.4%)	15 (71.4%)	61 (77.2%)
MRI Findings of NAFLD					< 0.001***
Yes	155 (24.8%)	135 (25.7%)	7 (33.3%)	13 (16.5%)
No	11 (1.8%)	6 (1.1%)	0 (0.0%)	5 (6.3%)
Missing	459 (73.4%)	384 (73.1%)	14 (66.7%)	61 (77.2%)
MRE Stage					< 0.001***
F0	22 (3.5%)	20 (3.8%)	2 (9.5%)	0 (0.0%)
F1	9 (1.4%)	1 (0.2%)	0 (0.0%)	8 (10.1%)
F2	8 (1.3%)	8 (1.5%)	0 (0.0%)	0 (0.0%)
F3	5 (0.8%)	2 (0.4%)	0 (0.0%)	3 (3.8%)
F4	6 (1.0%)	4 (0.8%)	1 (4.8%)	1 (1.3%)
Missing	575 (92.0%)	490 (93.3%)	18 (85.7%)	67 (84.8%)
Fibroscan CAP Steatosis Score	302.79 ± 65.30	302.96 ± 64.70	330.50 ± 56.13	296.65 ± 70.57	0.64
Fibroscan Steatosis Grade					0.75
None	20 (3.2%)	15 (2.9%)	0 (0.0%)	5 (6.3%)
S1	9 (1.4%)	5 (1.0%)	1 (4.8%)	3 (3.8%)
S2	10 (1.6%)	8 (1.5%)	0 (0.0%)	2 (2.5%)
S3	71 (11.4%)	52 (9.9%)	3 (14.3%)	16 (20.3%)
Missing	515 (82.4%)	445 (84.8%)	17 (81.0%)	53 (67.1%)
Fibroscan Fibrosis Score (kPa)	8.68 ± 7.50	9.46 ± 8.53	7.22 ± 3.14	6.49 ± 2.55	0.17
Fibrosis Score					0.86
F0	55 (8.8%)	38 (7.2%)	3 (14.3%)	14 (17.7%)
F1	26 (4.2%)	17 (3.2%)	1 (4.8%)	8 (10.1%)
F2	20 (3.2%)	13 (2.5%)	0 (0.0%)	7 (8.9%)
F3	26 (4.2%)	20 (3.8%)	0 (0.0%)	6 (7.6%)
F4	28 (4.5%)	21 (4.0%)	1 (4.8%)	6 (7.6%)
Missing	470 (75.2%)	416 (79.2%)	16 (76.2%)	38 (48.1%)
NAFLD Score	-0.72 ± 2.24	-0.47 ± 2.20	-1.62 ± 2.35	-2.17 ± 1.86	< 0.001***

Table: Table 1: Baseline Demographics of Study Groups

Disclosures:
Sindhura Kolli indicated no relevant financial relationships.
Sharnendra Sidhu indicated no relevant financial relationships.
Soonwook Hong indicated no relevant financial relationships.
Kristen Dammeyer indicated no relevant financial relationships.
Eli Rapoport indicated no relevant financial relationships.
Vlad Obsekov indicated no relevant financial relationships.
Gabrielle Maranga indicated no relevant financial relationships.
Holly Lofton: Eli Lily – Grant/Research Support. Novo Nordisk – Consultant, Speaker's Bureau.
Violeta Popov: Microtech – Grant/Research Support.

Sindhura Kolli, MD¹, Sharnendra Sidhu, MD¹, Soonwook Hong, MD¹, Kristen Dammeyer, ¹, Eli Rapoport, ¹, Vlad Obsekov, MD¹, Gabrielle Maranga, MPH¹, Holly F. Lofton, MD¹, Violeta Popov, MD, PhD². P0864 - Association Between Polycystic Ovarian Syndrome and Non-Alcoholic Fatty Liver Disease Fibrosis Severity: A Retrospective Cohort Analysis, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.