Preoperative ultrasound combined with routine blood tests in predicting the malignant risk of pancreatic cystic neoplasms

Xiuchao Wang; Junjin Wang; Xi Wei; Lihui Zhao; Bo Ni; Zekun Li; Chuntao Gao; Song Gao; Tiansuo Zhao; Jian Wang; Weidong Ma; Xiao Hu; Jihui Hao

doi:10.20892/j.issn.2095-3941.2022.0258

Article Figures & Data

Figures

Tables
Supplementary Materials

Download figure
Open in new tab
Download powerpoint
Figure 1
Screening of model variables. Variables were screened with Best Subsets Regression (BSR) (A, B; AIC = 151.3), Lasso regression (C, D; AIC = 152.27), and univariate regression and multiple logistic regression (AIC = 144.02).
Download figure
Open in new tab
Download powerpoint
Figure 2
(A) Multivariate analysis revealing that increased CA19-9, increased CEA, increased NLR, increased MCHC, non-O blood group, age > 55 years, cysts located in the head or neck of the pancreas, and an unclear cyst imaging border were independent predictors of PCN malignancy. (B) Nomogram for estimating the risk of malignancy in PCN. The nomogram is used to find the position of each variable on the corresponding axis, a vertical line is drawn to the points axis to obtain the value of the corresponding variable, the points are added for all variables, and then a line is drawn from the total points axis to determine the probability of a malignant tumor at the lower part of the nomogram.
Download figure
Open in new tab
Download powerpoint
Figure 3
ROC analysis of the malignancy prediction ability of the sum of scores. (A) Internal validation of the nomogram. (B) External validation of the nomogram. (C, D) Nomogram calibration in the model group (C) and the validation group (D). The horizontal axis indicates the risk of malignant PCN predicted by the nomogram, and the vertical axis is the probability of malignant PCN for a random sample of 1000 actual observations. The line from the lower left to the upper right corner of the plot region represents the reference line of the ideal prediction.
Download figure
Open in new tab
Download powerpoint
Figure 4
DCA curves for assessing the net benefit of model application in the model development cohort and validation cohort.

View popup

Table 1

Univariate logistic regression analysis based on preoperative data in the model cohort

Variable (%)	Level	Overall (%)	Benign (%)	Malignant (%)	OR (95% CI)	P
n	–	204	121	83	–	–
Age (years)	≤ 55	87 (42.6)	69 (57.0)	18 (21.7)	1.42 (1.25–1.61)	< 0.001*
	> 55	117 (57.4)	52 (43.0)	65 (78.3)	–	–
Gender	Female	122 (59.8)	84 (69.4)	38 (45.8)	1.27 (1.11–1.45)	0.001*
	Male	82 (40.2)	37 (30.6)	45 (54.2)	–	–
Location	Body/tail	111 (54.4)	84 (69.4)	27 (32.5)	1.43 (1.26–1.63)	< 0.001*
	Head/neck	93 (45.6)	37 (30.6)	56 (67.5)	–	–
Size (cm)	≤ 2.75	158 (77.5)	97 (80.2)	61 (73.5)	1.1 (0.93–1.29)	0.265
	> 2.75	46 (22.5)	24 (19.8)	22 (26.5)	–	–
Solid	No	79 (38.7)	62 (51.2)	17 (20.5)	1.37 (1.2–1.56)	< 0.001*
	Yes	125 (61.3)	59 (48.8)	66 (79.5)	–	–
Boundary	Clear	100 (49.0)	83 (68.6)	17 (20.5)	1.59 (1.41–1.79)	< 0.001*
	Blurred	104 (51.0)	38 (31.4)	66 (79.5)	–	–
Jaundice	No	187 (91.7)	120 (99.2)	67 (80.7)	1.79 (1.42–2.26)	< 0.001*
	Yes	17 (8.3)	1 (0.8)	16 (19.3)	–	–
Stomachache	No	109 (53.4)	75 (62.0)	34 (41.0)	1.23 (1.07–1.4)	0.003*
	Yes	95 (46.6)	46 (38.0)	49 (59.0)	–	–
Blood type	Type-O	51 (25.0)	35 (28.9)	16 (19.3)	1.13 (0.97–1.32)	0.119
	Non-O	153 (75.0)	86 (71.1)	67 (80.7)	–	–
CRP (mg/L)	≤ 1.29	45 (22.1)	11 (13.3)	34 (28.1)	1.23 (1.05–1.45)	0.012*
	> 1.29	159 (77.9)	72 (86.7)	87 (71.9)	–	–
RBC (×10¹²/L)	≤ 4.3	71 (34.8)	37 (30.6)	34 (41.0)	0.9 (0.78–1.03)	0.127
	> 4.3	133 (65.2)	84 (69.4)	49 (59.0)	–	–
HGB (g/L)	≤ 137.5	109 (53.4)	69 (57.0)	40 (48.2)	1.09 (0.95–1.25)	0.216
	> 137.5	95 (46.6)	52 (43.0)	43 (51.8)	–	–
HCT (%)	≤ 36.55	32 (15.7)	14 (11.6)	18 (21.7)	0.83 (0.69–1.00)	0.051
	> 36.55	172 (84.3)	107 (88.4)	65 (78.3)	–	–
MCV (fl)	≤ 90.65	93 (45.6)	60 (49.6)	33 (39.8)	1.1 (0.96–1.26)	0.168
	> 90.65	111 (54.4)	61 (50.4)	50 (60.2)	–	–
MCH (pg)	≤ 30.75	124 (60.8)	79 (65.3)	45 (54.2)	1.12 (0.97–1.28)	0.113
	> 30.75	80 (39.2)	42 (34.7)	38 (45.8)	–	–
MCHC (g/L)	≤ 341.5	174 (85.3)	109 (90.1)	65 (78.3)	1.25 (1.04–1.51)	0.020*
	> 341.5	30 (14.7)	12 (9.9)	18 (21.7)	–	–
RDW.CV (%)	≤ 13.4	164 (80.4)	107 (88.4)	57 (68.7)	1.35 (1.15–1.6)	< 0.001*
	> 13.4	40 (19.6)	14 (11.6)	26 (31.3)	–	–
HR	≤ 2.875	43 (21.1)	16 (13.2)	27 (32.5)	0.76 (0.64–0.89)	0.001*
	> 2.875	161 (78.9)	105 (86.8)	56 (67.5)	–	–
RETR (%)	≤ 0.985	24 (11.8)	11 (9.1)	13 (15.7)	0.86 (0.7–1.06)	0.154
	> 0.985	180 (88.2)	110 (90.9)	70 (84.3)	–	–
RET (×10⁹/L)	≤ 82.4	158 (77.5)	90 (74.4)	68 (81.9)	0.90 (0.77–1.06)	0.207
	> 82.4	46 (22.5)	31 (25.6)	15 (18.1)	–	–
WBC (×10⁹/L)	≤ 6.16	124 (60.8)	81 (66.9)	43 (51.8)	1.17 (1.02–1.34)	0.030*
	> 6.16	80 (39.2)	40 (33.1)	40 (48.2)	–	–
NEUT (×10⁹/L)	≤ 2.6	83 (40.7)	58 (47.9)	25 (30.1)	1.19 (1.04–1.37)	0.011*
	> 2.6	121 (59.3)	63 (52.1)	58 (69.9)	–	–
LYMPH (×10⁹/L)	≤ 1.905	132 (64.7)	71 (58.7)	61 (73.5)	0.86 (0.74–0.98)	0.030*
	> 1.905	72 (35.3)	50 (41.3)	22 (26.5)	–	–
MONO (×10⁹/L)	≤ 0.535	119 (58.3)	79 (65.3)	40 (48.2)	1.19 (1.03–1.36)	0.015*
	> 0.535	85 (41.7)	42 (34.7)	43 (51.8)	–	–
PLT (×10⁹/L)	≤ 207.5	68 (33.3)	34 (28.1)	34 (41.0)	0.87 (0.75–1.00)	0.056
	> 207.5	136 (66.7)	87 (71.9)	49 (59.0)	–	–
MPV (fl)	≤ 11.25	150 (73.5)	99 (81.8)	51 (61.4)	1.29 (1.11–1.5)	0.001*
	> 11.25	54 (26.5)	22 (18.2)	32 (38.6)	–	–
P-LCR (%)	≤ 37.85	164 (80.4)	107 (88.4)	57 (68.7)	1.35 (1.15–1.6)	< 0.001*
	> 37.85	40 (19.6)	14 (11.6)	26 (31.3)	–	–
PLR	≤ 196.47	23 (11.3)	10 (8.3)	13 (15.7)	0.84 (0.68–1.03)	0.102
	> 196.47	181 (88.7)	111 (91.7)	70 (84.3)	–	–
NLR	≤ 1.79	92 (45.1)	67 (55.4)	25 (30.1)	1.28 (1.12–1.46)	< 0.001*
	> 1.79	112 (54.9)	54 (44.6)	58 (69.9)	–	–
LMR	≤ 3.64	61 (29.9)	22 (18.2)	39 (47.0)	0.72 (0.62–0.83)	< 0.001*
	> 3.64	143 (70.1)	99 (81.8)	44 (53.0)	–	–
CA19-9 (U/mL)	−	146 (71.6)	109 (90.1)	37 (44.6)	1.72 (1.51–1.95)	< 0.001*
	+	58 (28.4)	12 (9.9)	46 (55.4)	–	–
CA242 (U/mL)	−	170 (83.3)	117 (96.7)	53 (63.9)	1.77 (1.5–2.08)	< 0.001*
	+	34 (16.7)	4 (3.3)	30 (36.1)	–	–
CEA (μg/L)	−	184 (90.2)	120 (99.2)	64 (77.1)	1.83 (1.48–2.26)	< 0.001*
	+	20 (9.8)	1 (0.8)	19 (22.9)	–	–

^*P < 0.05 was considered statistically significant. CRP, C-reactive protein; RBC, red blood cell; HGB, hemoglobin; HCT, red blood cell specific volume; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; RDW.CV, red blood cell volume distribution width; HR, hemoglobin-red blood cell distribution width ratio; RETR, reticulocyte ratio; RET, reticulocyte; WBC, peripheral white blood cells; NEUT, neutrophils; LYMPH, lymphocyte; MONO, monocyte; PLT, platelet; MPV, mean platelet volume; P-LCR, platelet-larger cell ratio; PLR, lymphocyte-monocyte ratio; NLR, neutrophil-lymphocyte ratio; LMR, lymphocyte-monocyte ratio; CA19-9, carbohydrate antigen 19-9; CA242, carbohydrate chain antigen 242; CEA, carcinoembryonic antigen; OR, odds ratio; CI, confidence interval; +, CA199 > 37 U/ml, CA242 > 20 U/ml, CEA > 5 μg/L.

View popup

Table 2

Differences between the development cohort and validation cohort

Variable	Level	Overall (%)	Development cohort	Validation cohort	P
No. patients		290	204	86
Age (years)	≤ 55	117 (40.3)	87	30	0.218
	> 55	173 (59.7)	117	56
Location	Body/tail	164 (56.6)	111	53	0.258
	Head/neck	126 (43.4)	93	33
Boundary	Clear	144 (49.7)	100	44	0.739
	Blurred	146 (50.3)	104	42
Blood type	Type-O	79 (27.2)	51	28	0.187
	Non-O	211 (72.8)	153	58
MCHC (g/L)	≤ 341.5	241 (83.1)	174	67	0.125
	> 341.5	49 (16.9)	30	19
NLR	≤ 1.79	132 (45.5)	92	40	0.825
	> 1.79	158 (54.5)	112	46
CA19-9 (U/mL)	≤ 37	209 (72.1)	146	63	0.770
	> 37	81 (27.9)	58	23
CEA (μg/L)	≤ 5	256 (88.3)	184	72	0.117
	> 5	34 (11.7)	20	14

View popup

Table 3

Risk stratification and restratification

Imaging and tumor marker model	Nomogram risk model		Total	Reclassified as higher risk (%)	Reclassified as lower risk (%)
Risk	< 0.5	≥ 0.5	Total	Reclassified as higher risk (%)	Reclassified as lower risk (%)
< 0.5
No. of patients	117	33	150	33 (22%)	NA
No. of benign	105	7	112	7 (6.25%)	NA
No. of malignant	12	26	38	26 (68.4%)	NA
≥ 0.5
No. of patients	4	50	54	NA	4 (7.4%)
No. of benign	3	6	9	NA	3 (33.3%)
No. of malignant	1	44	45	NA	1 (2.2%)
Total
No. of patients	121	83	204	33 (22%)	4 (7.4%)
No. of benign	108	13	121	7 (6.25%)	3 (33.3%)
No. of malignant	13	70	83	26 (68.4%)	1 (2.2%)