Predictors of pain medication after liver biopsy

Author

Lu Mao

Published

June 30, 2025

Statistical analysis

Categorical variables are summarized as counts and percentages, continuous variables as median and interquartile range (IQR). The treatment group is compared to the control group using Fisher’s exact test for categorical variables and Wilcoxon rank-sum test for continuous variables. A multiple logistic regression model is fitted to assess the association of predictors with treatment vs control. P values are two-sided and considered significant at the 0.05 level. The analysis is performed using R version 4.3.2 (R Core Team, 2025).

Decriptive anaysis

The initial dataset contains 669 subjects, of which 96 (14.3%) are excluded. The remaining 573 subjects meet inclusion criteria Exclude == 1. Subjects demographics, lab values, pathology and imaging data are tabulated by treatment vs control in Table 1, Table 2, and Table 3.

Demographics and labs

Table 1: Patient characteristics and lab values by treatment.
Characteristic Overall, N = 5731 Treatment, N = 2921 Control, N = 2811 p-value2
Performed by



    Radiology 573 (100%) 292 (100%) 281 (100%)
Department


0.5
    Outpatient 556 (97%) 285 (98%) 271 (96%)
    Inpatient 14 (2.4%) 5 (1.7%) 9 (3.2%)
    Other 3 (0.5%) 2 (0.7%) 1 (0.4%)
Age (years) 56 (41, 67) 51 (36, 64) 60 (49, 69) <0.001
Sex


<0.001
    Male 233 (41%) 86 (29%) 147 (52%)
    Female 340 (59%) 206 (71%) 134 (48%)
Ketorolac 268 (47%) 268 (92%) 0 (0%) <0.001
dose_amt_tor


<0.001
    0 305 (53%) 24 (8.2%) 281 (100%)
    15 114 (20%) 114 (39%) 0 (0%)
    30 148 (26%) 148 (51%) 0 (0%)
    45 3 (0.5%) 3 (1.0%) 0 (0%)
    60 3 (0.5%) 3 (1.0%) 0 (0%)
dose_nbr_tor


<0.001
    0 305 (53%) 24 (8.2%) 281 (100%)
    1 249 (43%) 249 (85%) 0 (0%)
    2 19 (3.3%) 19 (6.5%) 0 (0%)
dosage_tor


>0.9
    (0, 15] 128 (48%) 128 (48%) 0 (NA%)
    (15, 30] 137 (51%) 137 (51%) 0 (NA%)
    > 30 3 (1.1%) 3 (1.1%) 0 (NA%)
    Unknown 305 24 281
Hydromorphone 44 (7.7%) 44 (15%) 0 (0%) <0.001
dose_amt_dil


<0.001
    0 529 (92%) 248 (85%) 281 (100%)
    (0.2, 0.5] 32 (5.6%) 32 (11%) 0 (0%)
    (0.5, 1.0] 8 (1.4%) 8 (2.7%) 0 (0%)
    > 1.0 4 (0.7%) 4 (1.4%) 0 (0%)
dose_nbr_dil


<0.001
    0 529 (92%) 248 (85%) 281 (100%)
    1 36 (6.3%) 36 (12%) 0 (0%)
    2 6 (1.0%) 6 (2.1%) 0 (0%)
    3 1 (0.2%) 1 (0.3%) 0 (0%)
    4 1 (0.2%) 1 (0.3%) 0 (0%)
dosage_dil 0.30 (0.20, 0.50) 0.30 (0.20, 0.50) NA (NA, NA)
    Unknown 530 249 281
Received both 22 (3.8%) 22 (7.5%) 0 (0%) <0.001
Other_Med_Given


0.042
    1.0 8 (29%) 5 (20%) 3 (100%)
    2.0 18 (64%) 18 (72%) 0 (0%)
    3.0 2 (7.1%) 2 (8.0%) 0 (0%)
    Unknown 545 267 278
pre_AST 40 (25, 70) 38 (24, 66) 43 (27, 75) 0.078
    Unknown 172 84 88
pre_AST2


0.13
    < 34 163 (41%) 92 (44%) 71 (37%)
    ≥ 34 238 (59%) 116 (56%) 122 (63%)
    Unknown 172 84 88
pre_ALT 42 (24, 95) 38 (22, 90) 48 (25, 100) 0.069
    Unknown 168 81 87
pre_ALT2


0.067
    < 55 234 (58%) 131 (62%) 103 (53%)
    ≥ 55 171 (42%) 80 (38%) 91 (47%)
    Unknown 168 81 87
pre_ALKP 110 (81, 185) 105 (78, 156) 119 (85, 232) 0.012
    Unknown 178 86 92
pre_ALKP2


0.004
    < 150 258 (65%) 148 (72%) 110 (58%)
    ≥ 150 137 (35%) 58 (28%) 79 (42%)
    Unknown 178 86 92
pre_INR 1.00 (1.00, 1.10) 1.00 (1.00, 1.10) 1.00 (1.00, 1.10) 0.2
    Unknown 60 37 23
pre_INR2


0.15
    < 1.2 424 (83%) 217 (85%) 207 (80%)
    ≥ 1.2 89 (17%) 38 (15%) 51 (20%)
    Unknown 60 37 23
Anticoag_Type


<0.001
    None 413 (72%) 236 (81%) 177 (63%)
    Blood thinner 35 (6.1%) 18 (6.2%) 17 (6.0%)
    AntiPLT 116 (20%) 36 (12%) 80 (28%)
    Blood thinner + AntiPLT 9 (1.6%) 2 (0.7%) 7 (2.5%)
pre_PLT 240 (179, 293) 244 (194, 299) 227 (166, 287) 0.061
    Unknown 85 42 43
Hgb_drawn_postbx 33 (5.8%) 9 (3.1%) 24 (8.5%) 0.005
Hgb_drop 7 (19%) 2 (18%) 5 (20%) >0.9
    Unknown 537 281 256
Hypotension 5 (0.9%) 2 (0.7%) 3 (1.1%) 0.7
    Unknown 25 6 19
Level_of_Care


0.3
    1 9 (1.6%) 7 (2.4%) 2 (0.7%)
    2 14 (2.4%) 8 (2.7%) 6 (2.1%)
    3 11 (1.9%) 4 (1.4%) 7 (2.5%)
    4 530 (92%) 270 (92%) 260 (93%)
    5 9 (1.6%) 3 (1.0%) 6 (2.1%)
Indication_code


<0.001
    Liver function abnormality 210 (37%) 122 (42%) 88 (31%)
    Liver lesion evaluation 335 (58%) 168 (58%) 167 (59%)
    Liver transplant elevated LFTs 28 (4.9%) 2 (0.7%) 26 (9.3%)
Death_within_30d 8 (1.4%) 3 (1.0%) 5 (1.8%) 0.5
Biopsy_target


<0.001
    Target 327 (57%) 163 (56%) 164 (58%)
    Random 219 (38%) 127 (43%) 92 (33%)
    Txp 27 (4.7%) 2 (0.7%) 25 (8.9%)
1 n (%); Median (IQR)
2 Fisher’s exact test; Wilcoxon rank sum test; Pearson’s Chi-squared test

Pathology

Some levels may need to be combined.

Table 2: Patient pathodology by treatment.
Characteristic Overall, N = 5731 Treatment, N = 2921 Control, N = 2811 p-value2
Pathology_Mets


0.056
    Mets 226 (40%) 104 (36%) 122 (44%)
    No Mets 345 (60%) 187 (64%) 158 (56%)
    Unknown 2 1 1
Pathology_TM


0.063
    No non-cirrhotic liver injury 445 (78%) 236 (81%) 209 (75%)
    Non-cirrhotic liver injury 126 (22%) 55 (19%) 71 (25%)
    Unknown 2 1 1
Pathology_SHF


0.5
    0 373 (65%) 186 (64%) 187 (67%)
    1 57 (10.0%) 31 (11%) 26 (9.3%)
    2/3 44 (7.7%) 26 (8.9%) 18 (6.4%)
    4 33 (5.8%) 20 (6.9%) 13 (4.6%)
    5 39 (6.8%) 18 (6.2%) 21 (7.5%)
    6 25 (4.4%) 10 (3.4%) 15 (5.4%)
    Unknown 2 1 1
Pathology_Biliary


0.4
    0 503 (88%) 258 (89%) 245 (88%)
    1,5 25 (4.4%) 9 (3.1%) 16 (5.7%)
    2-4 16 (2.8%) 10 (3.4%) 6 (2.1%)
    6-7 26 (4.6%) 14 (4.8%) 12 (4.3%)
    8 1 (0.2%) 0 (0%) 1 (0.4%)
    Unknown 2 1 1
Pathology_BenignOther


0.033
    0 490 (86%) 237 (81%) 253 (90%)
    1-2,7-8 35 (6.1%) 24 (8.2%) 11 (3.9%)
    3 11 (1.9%) 8 (2.7%) 3 (1.1%)
    4 29 (5.1%) 19 (6.5%) 10 (3.6%)
    5-6 6 (1.1%) 3 (1.0%) 3 (1.1%)
    Unknown 2 1 1
Pathology_Txp


0.001
    0 557 (98%) 290 (100%) 267 (95%)
    1 5 (0.9%) 0 (0%) 5 (1.8%)
    2 4 (0.7%) 0 (0%) 4 (1.4%)
    3 5 (0.9%) 1 (0.3%) 4 (1.4%)
    Unknown 2 1 1
1 n (%)
2 Pearson’s Chi-squared test; Fisher’s exact test

Imaging and others

Table 3: Patient imaging results and others by treatment.
Characteristic Overall, N = 5731 Treatment, N = 2921 Control, N = 2811 p-value2
US_post


0.7
    1 534 (93%) 270 (92%) 264 (94%)
    2 10 (1.7%) 7 (2.4%) 3 (1.1%)
    3 24 (4.2%) 12 (4.1%) 12 (4.3%)
    4 4 (0.7%) 2 (0.7%) 2 (0.7%)
    5 1 (0.2%) 1 (0.3%) 0 (0%)
Post_US_image


0.5
    1 39 (6.8%) 22 (7.5%) 17 (6.0%)
    2 534 (93%) 270 (92%) 264 (94%)
CT_postBx


0.029
    1 553 (97%) 277 (95%) 276 (98%)
    2 20 (3.5%) 15 (5.1%) 5 (1.8%)
CT_Finding 4 (22%) 3 (21%) 1 (25%) >0.9
    Unknown 555 278 277
CT_Finding_type


0.7
    1 10 (77%) 7 (78%) 3 (75%)
    2 1 (7.7%) 1 (11%) 0 (0%)
    3 1 (7.7%) 0 (0%) 1 (25%)
    4 1 (7.7%) 1 (11%) 0 (0%)
    Unknown 560 283 277
SecondLook_US_Finding 3 (7.7%) 3 (9.4%) 0 (0%) >0.9
    Unknown 534 260 274
PainScore_10p


<0.001
    Median (IQR) 0 (0, 6) 6 (3, 8) 0 (0, 0)
    Mean (SD) 846 (2,782) 1,000 (2,999) 712 (2,575)
    Unknown 99 71 28
PainScore_Y_N 284 (53%) 236 (86%) 48 (18%) <0.001
    Unknown 37 16 21
BiopsyDevice_code


0.2
    1 488 (97%) 225 (95%) 263 (98%)
    2 6 (1.2%) 3 (1.3%) 3 (1.1%)
    3 8 (1.6%) 6 (2.5%) 2 (0.7%)
    4 2 (0.4%) 2 (0.8%) 0 (0%)
    Unknown 69 56 13
PassLength_CM 2.30 (2.30, 3.30) 2.40 (2.20, 3.30) 2.30 (2.30, 3.30) 0.4
    Unknown 24 14 10
Biopsy_gauge



    18 560 (100%) 285 (100%) 275 (100%)
    Unknown 13 7 6
Biopsy_side


<0.001
    1.0 304 (53%) 179 (61%) 125 (44%)
    2.0 194 (34%) 83 (28%) 111 (40%)
    3.0 2 (0.3%) 0 (0%) 2 (0.7%)
    4.0 3 (0.5%) 1 (0.3%) 2 (0.7%)
    9999.0 64 (11%) 28 (9.6%) 36 (13%)
    Both 6 (1.0%) 1 (0.3%) 5 (1.8%)
Num_pass


0.13
    Mean (SD) 2.00 (1.13) 2.04 (1.10) 1.95 (1.16)
    Unknown 2 1 1
Time_to_med 0.024 (0.014, 0.036) 0.024 (0.014, 0.036) 0.015 (0.015, 0.015) 0.5
    Unknown 285 5 280
1 n (%); Median (IQR)
2 Fisher’s exact test; Pearson’s Chi-squared test; Wilcoxon rank sum test

Multivariate analysis

A multiple logistic regression model for the probability of control (vs treatment) is fitted against age, sex, pre-ALKP, anticoagulation type, post-biopsy Hgb drawn, indication code, biopsy target, pathology benign other, pathology transplant, and biopsy side. The results are summarized in Table 4.

Table 4: Multivariate analysis of predictors of pain medication after liver biopsy.
Characteristic OR1 95% CI1 p-value
Age (years) 1.04 1.02, 1.06 <0.001
Sex

<0.001
    Male
    Female 0.32 0.19, 0.52
pre_ALKP2

0.2
    < 150
    ≥ 150 1.41 0.85, 2.32
Anticoag_Type

0.4
    None
    Blood thinner 0.87 0.33, 2.25
    AntiPLT 1.79 0.88, 3.73
    Blood thinner + AntiPLT 0.99 0.10, 10.3
Hgb_drawn_postbx

0.066
    No
    Yes 0.17 0.01, 1.11
Indication_code

0.14
    Liver function abnormality
    Liver lesion evaluation 1.48 0.31, 6.78
    Liver transplant elevated LFTs 147,835,242 0.00, NA
Biopsy_target

0.8
    Target
    Random 1.39 0.29, 6.45
    Txp 0.00

Pathology_BenignOther

0.040
    0
    1-2, 7-8 0.29 0.06, 1.12
    3 0.30 0.04, 1.52
    4 0.28 0.07, 0.89
    5-6 2.29 0.24, 52.4
Pathology_Txp 0.70 0.25, 2.36 0.5
Biopsy_side

0.005
    Right
    Left 1.88 1.12, 3.16
    Central/Bilateral 38,424,716 0.00, NA
    Missing 1.18 0.54, 2.56
1 OR = Odds Ratio, CI = Confidence Interval