Predictors of pain medication after liver biopsy

Author

Lu Mao

Statistical analysis

Placeholder for description of statistical methods.

Decriptive anaysis

The initial dataset contains 450 subjects, of which 47 (10.4%) are excluded. The remaining 403 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 = 4031 Treatment, N = 2031 Control, N = 2001 p-value2
Performed by


0.029
    Radiology 398 (99%) 203 (100%) 195 (98%)
    Transplant 5 (1.2%) 0 (0%) 5 (2.5%)
Department


0.4
    Outpatient 389 (97%) 198 (98%) 191 (96%)
    Inpatient 10 (2.5%) 3 (1.5%) 7 (3.5%)
    Other 4 (1.0%) 2 (1.0%) 2 (1.0%)
Age (years) 59 (43, 68) 51 (38, 66) 63 (51, 70) <0.001
Sex


<0.001
    Male 163 (40%) 59 (29%) 104 (52%)
    Female 240 (60%) 144 (71%) 96 (48%)
Ketorolac 187 (46%) 187 (92%) 0 (0%) <0.001
dose_amt_tor


<0.001
    0 216 (54%) 16 (7.9%) 200 (100%)
    15 88 (22%) 88 (43%) 0 (0%)
    30 93 (23%) 93 (46%) 0 (0%)
    45 3 (0.7%) 3 (1.5%) 0 (0%)
    60 3 (0.7%) 3 (1.5%) 0 (0%)
dose_nbr_tor


<0.001
    0 216 (54%) 16 (7.9%) 200 (100%)
    1 171 (42%) 171 (84%) 0 (0%)
    2 16 (4.0%) 16 (7.9%) 0 (0%)
dosage_tor


>0.9
    7.5 2 (1.1%) 2 (1.1%) 0 (NA%)
    15 97 (52%) 97 (52%) 0 (NA%)
    22.5 3 (1.6%) 3 (1.6%) 0 (NA%)
    30 82 (44%) 82 (44%) 0 (NA%)
    60 3 (1.6%) 3 (1.6%) 0 (NA%)
    Unknown 216 16 200
Hydromorphone 29 (7.2%) 29 (14%) 0 (0%) <0.001
dose_amt_dil


<0.001
    0 374 (93%) 174 (86%) 200 (100%)
    0.2 16 (4.0%) 16 (7.9%) 0 (0%)
    0.4 3 (0.7%) 3 (1.5%) 0 (0%)
    0.5 5 (1.2%) 5 (2.5%) 0 (0%)
    0.6 3 (0.7%) 3 (1.5%) 0 (0%)
    0.8 1 (0.2%) 1 (0.5%) 0 (0%)
    2 mg 1 (0.2%) 1 (0.5%) 0 (0%)
dose_nbr_dil


<0.001
    0 374 (93%) 174 (86%) 200 (100%)
    1 22 (5.5%) 22 (11%) 0 (0%)
    2 5 (1.2%) 5 (2.5%) 0 (0%)
    3 1 (0.2%) 1 (0.5%) 0 (0%)
    4 1 (0.2%) 1 (0.5%) 0 (0%)
dosage_dil


>0.9
    0.1 2 (7.1%) 2 (7.1%) 0 (NA%)
    0.15 1 (3.6%) 1 (3.6%) 0 (NA%)
    0.2 16 (57%) 16 (57%) 0 (NA%)
    0.3 1 (3.6%) 1 (3.6%) 0 (NA%)
    0.4 2 (7.1%) 2 (7.1%) 0 (NA%)
    0.5 5 (18%) 5 (18%) 0 (NA%)
    0.8 1 (3.6%) 1 (3.6%) 0 (NA%)
    Unknown 375 175 200
Received both 14 (3.5%) 14 (6.9%) 0 (0%) <0.001
Other_Med_Given


0.13
    Acetaminophen 8 (42%) 5 (31%) 3 (100%)
    Fentanyl 10 (53%) 10 (63%) 0 (0%)
    Oxycodone 1 (5.3%) 1 (6.3%) 0 (0%)
    Unknown 384 187 197
pre_AST 39 (24, 69) 36 (23, 66) 42 (25, 73) 0.15
    Unknown 120 57 63
pre_AST2


0.051
    < 34 120 (42%) 70 (48%) 50 (36%)
    ≥ 34 163 (58%) 76 (52%) 87 (64%)
    Unknown 120 57 63
pre_ALT 41 (22, 87) 36 (21, 81) 51 (24, 97) 0.063
    Unknown 117 54 63
pre_ALT2


0.012
    < 55 168 (59%) 98 (66%) 70 (51%)
    ≥ 55 118 (41%) 51 (34%) 67 (49%)
    Unknown 117 54 63
pre_ALKP 116 (81, 193) 107 (78, 160) 132 (85, 235) 0.010
    Unknown 122 58 64
pre_ALKP2


0.008
    < 150 177 (63%) 102 (70%) 75 (55%)
    ≥ 150 104 (37%) 43 (30%) 61 (45%)
    Unknown 122 58 64
pre_INR 1.00 (1.00, 1.10) 1.00 (1.00, 1.10) 1.00 (1.00, 1.10) 0.4
    Unknown 45 28 17
pre_INR2


0.6
    < 1.2 305 (85%) 151 (86%) 154 (84%)
    ≥ 1.2 53 (15%) 24 (14%) 29 (16%)
    Unknown 45 28 17
Anticoag_Type


<0.001
    None 278 (69%) 160 (79%) 118 (59%)
    Blood thinner 30 (7.4%) 13 (6.4%) 17 (8.5%)
    AntiPLT 88 (22%) 29 (14%) 59 (30%)
    Blood thinner + AntiPLT 7 (1.7%) 1 (0.5%) 6 (3.0%)
pre_PLT 242 (182, 299) 251 (197, 305) 228 (175, 288) 0.070
    Unknown 67 31 36
Hgb_drawn_postbx 29 (7.2%) 6 (3.0%) 23 (12%) <0.001
Hgb_drop 6 (19%) 1 (14%) 5 (21%) >0.9
    Unknown 372 196 176
Hypotension 3 (0.7%) 1 (0.5%) 2 (1.0%) 0.6
    Unknown 2 0 2
Level_of_Care


0.7
    1 3 (0.7%) 2 (1.0%) 1 (0.5%)
    2 9 (2.2%) 4 (2.0%) 5 (2.5%)
    3 9 (2.2%) 3 (1.5%) 6 (3.0%)
    4 379 (94%) 193 (95%) 186 (93%)
    5 3 (0.7%) 1 (0.5%) 2 (1.0%)
Indication_code


<0.001
    Liver function abnormality 124 (31%) 75 (37%) 49 (25%)
    Liver lesion evaluation 250 (62%) 126 (62%) 124 (62%)
    Liver transplant elevated LFTs 29 (7.2%) 2 (1.0%) 27 (14%)
Death_within_30d 6 (1.5%) 2 (1.0%) 4 (2.0%) 0.4
Biopsy_target


<0.001
    Target 249 (62%) 124 (61%) 125 (63%)
    Random 126 (31%) 77 (38%) 49 (25%)
    Txp 28 (6.9%) 2 (1.0%) 26 (13%)
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 = 4031 Treatment, N = 2031 Control, N = 2001 p-value2
Pathology_Mets


0.2
    0 231 (57%) 123 (61%) 108 (54%)
    11 13 (3.2%) 8 (3.9%) 5 (2.5%)
    Otherwise 159 (39%) 72 (35%) 87 (44%)
Pathology_TM


0.3
    0 322 (80%) 166 (82%) 156 (78%)
    1-9 81 (20%) 37 (18%) 44 (22%)
Pathology_SHF


0.081
    0 279 (69%) 141 (69%) 138 (69%)
    1-2 49 (12%) 30 (15%) 19 (9.5%)
    3-4 34 (8.4%) 18 (8.9%) 16 (8.0%)
    5-6 41 (10%) 14 (6.9%) 27 (14%)
Pathology_Biliary


0.5
    0 348 (86%) 178 (88%) 170 (85%)
    1-5, 8 34 (8.4%) 14 (6.9%) 20 (10%)
    6-7 21 (5.2%) 11 (5.4%) 10 (5.0%)
Pathology_BenignOther


0.058
    0 342 (85%) 163 (80%) 179 (90%)
    1-2, 7-8 20 (5.0%) 14 (6.9%) 6 (3.0%)
    3-4 35 (8.7%) 23 (11%) 12 (6.0%)
    5-6 6 (1.5%) 3 (1.5%) 3 (1.5%)
Pathology_Txp


<0.001
    0 387 (96%) 202 (100%) 185 (93%)
    1 5 (1.2%) 0 (0%) 5 (2.5%)
    2 4 (1.0%) 0 (0%) 4 (2.0%)
    3 7 (1.7%) 1 (0.5%) 6 (3.0%)
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 = 4031 Treatment, N = 2031 Control, N = 2001 p-value2
US_post


0.3
    1 378 (94%) 189 (93%) 189 (95%)
    2 7 (1.7%) 6 (3.0%) 1 (0.5%)
    3 14 (3.5%) 6 (3.0%) 8 (4.0%)
    4 3 (0.7%) 1 (0.5%) 2 (1.0%)
    5 1 (0.2%) 1 (0.5%) 0 (0%)
Post_US_image


0.6
    1 25 (6.2%) 14 (6.9%) 11 (5.5%)
    2 378 (94%) 189 (93%) 189 (95%)
CT_postBx


0.4
    N 388 (96%) 194 (96%) 194 (97%)
    Y 15 (3.7%) 9 (4.4%) 6 (3.0%)
CT_Finding 4 (29%) 3 (33%) 1 (20%) >0.9
    Unknown 389 194 195
CT_Finding_type


0.8
    1 11 (79%) 7 (78%) 4 (80%)
    2 1 (7.1%) 1 (11%) 0 (0%)
    3 1 (7.1%) 0 (0%) 1 (20%)
    4 1 (7.1%) 1 (11%) 0 (0%)
    Unknown 389 194 195
SecondLook_US_Finding



    No 20 (100%) 18 (100%) 2 (100%)
    Unknown 383 185 198
PainScore_10p


<0.001
    Median (IQR) 0.00 (0.00, 5.00) 5.00 (0.00, 7.00) 0.00 (0.00, 0.00)
    Mean (SD) 2.18 (3.16) 4.53 (3.15) 0.37 (1.61)
    Unknown 101 72 29
PainScore_Y_N


<0.001
    No 196 (50%) 34 (17%) 162 (82%)
    Yes 195 (50%) 160 (82%) 35 (18%)
    Y 1 (0.3%) 1 (0.5%) 0 (0%)
    Unknown 11 8 3
BiopsyDevice_code


0.6
    1 347 (96%) 161 (95%) 186 (97%)
    2 6 (1.7%) 3 (1.8%) 3 (1.6%)
    3 5 (1.4%) 3 (1.8%) 2 (1.0%)
    4 2 (0.6%) 2 (1.2%) 0 (0%)
    Unknown 43 34 9
PassLength_CM 2.30 (1.30, 3.30) 2.30 (1.30, 3.30) 2.30 (2.20, 3.30) 0.8
    Unknown 13 7 6
Biopsy_gauge



    18 393 (100%) 198 (100%) 195 (100%)
    Unknown 10 5 5
Biopsy_side


<0.001
    1 209 (61%) 126 (71%) 83 (51%)
    2 123 (36%) 49 (28%) 74 (45%)
    3 2 (0.6%) 0 (0%) 2 (1.2%)
    4 1 (0.3%) 1 (0.6%) 0 (0%)
    Both 6 (1.8%) 1 (0.6%) 5 (3.0%)
    Unknown 62 26 36
Num_pass


0.027
    Mean (SD) 2.07 (1.11) 2.17 (1.12) 1.97 (1.10)
    Unknown 1 1 0
Time_to_med 0.022 (0.013, 0.033) 0.022 (0.013, 0.032) 0.044 (0.029, 0.059) 0.5
    Unknown 203 5 198
1 n (%); Median (IQR)
2 Fisher’s exact test; Pearson’s Chi-squared test; Wilcoxon rank sum test

Multivariate analysis

Predictors to be included in multiple logistic regression for treatment vs control:

  1. Given \(n=\) 403, total number of predictors should be < 20;
  2. Include significant factors with few missing values (consider medical relevance as well).
  3. Variables with too many categories may need to be grouped.