Prediction and stratification for the surgical adverse events after minimally invasive esophagectomy: A two-center retrospective study

Table 1 Compared different baseline information between non-surgical adverse events and surgical adverse events patients, n (%)

	Train set (n = 549)			Validation set (n = 198)
	Non-SAE (n = 493)	SAE (n = 56)	P value	Non-SAE (n = 171)	SAE (n = 27)	P value
Age, mean ± SD	61.11 ± 7.32	65.77 ± 5.43	< 0.001a	61.87 ± 7.28	64.15 ± 7.17	0.131
Gender (male)	397 (80.5)	40 (71.4)	0.109	130 (76.0)	19 (70.4)	0.527
BMI, mean ± SD	22.10 ± 2.76	21.65 ± 3.09	0.252	21.81 ± 2.76	22.11 ± 2.92	0.604
Smoke (+)	76 (53.5)	115 (59.9)	0.224	102 (59.6)	16 (59.3)	0.969
History of alcohol consumption (+)	43 (30.3)	61 (31.8)	0.771	42 (24.6)	8 (29.6)	0.573
Diabetes (+)	35 (7.1)	4 (7.1)	0.990	10 (5.8)	2 (7.4)	0.752
HBP (+)	99 (20.1)	15 (26.8)	0.241	32 (18.7)	3 (11.1)	0.336
CHD (+)	35 (7.1)	6 (10.7)	0.329	10 (5.8)	2 (7.4)	0.752
COPD (+)	11 (2.2)	7 (12.5)	< 0.001a	3 (1.8)	4 (14.8)	0.007a
Tumor location			0.057			0.856
Upper thoracic	39 (7.9)	1 (1.8)		11 (6.4)	1 (3.7)
Middle thoracic	194 (39.4)	30 (53.6)		69 (40.4)	11 (40.7)
Lower thoracic	260 (52.7)	25 (44.6)		91 (53.2)	15 (55.6)
Neoadjuvant therapy (+)	277 (56.2)	25 (44.6)	0.100	83 (48.5)	14 (51.9)	0.749
pT			0.857			0.117
0	55 (11.2)	8 (14.3)		17 (9.9)	4 (14.8)
1	154 (31.2)	17 (30.4)		55 (32.2)	4 (14.8)
2	77 (15.6)	7 (12.5)		30 (17.5)	9 (33.3)
3	207 (42.0)	24 (42.9)		69 (40.4)	10 (37.0)
4	0 (0)	0 (0)		0 (0)	0 (0)
pN			0.746			0.509
0	285 (57.8)	36 (64.3)		99 (57.9)	18 (66.7)
1	122 (24.7)	13 (23.2)		41 (24.0)	3 (11.1)
2	69 (14.0)	6 (10.7)		27 (15.8)	5 (18.5)
3	17 (3.4)	1 (1.8)		4 (2.3)	1 (3.7)

^aP < 0.05.

SAE: Surgical adverse events; BMI: Body mass index; HBP: High blood pressure; CHD: Coronary heart disease; COPD: Chronic obstructive pulmonary disease.

Table 2 Compared different short-term outcomes between subgroups of train set, n (%)

	Train set (n = 549)
	Non-SAE (n = 493)	SAE (n = 56)	P value
Pulmonary infection (+)	126 (25.6)	22 (39.3)	0.028
Blood transfusion (+)	0 (0.0)	12 (21.4)	< 0.001
ICU management (+)	0 (0.0)	37 (66.1)	< 0.001
Anastomotic leak (+)	31 (6.3)	45 (80.4)	< 0.001
Re-operation (+)	0 (0.0)	3 (5.4)	< 0.0012
Duration of chest tube drainage, medium (IQR)	8 (4)	13 (15)	< 0.0011
ICU stay time, medium (IQR)	0 (0)	7 (9)	< 0.0011
Death within 30-day	0 (0.0)	0 (0.0)	/

¹Compared using Kruskal-Wallis rank-sum test.

²Adjusted P values.

IQR: Interquartile range; ICU: Intensive care unit; SAE: Surgical adverse event.

Table 3 Univariate and multivariate analysis of influencing factors

Variable	Univariate			Multivariate
	OR	95%CI	P value	OR	95%CI	P value
Age	1.10	1.05-1.14	< 0.001a	1.03	0.97-1.08	0.339
History of alcohol consumption			0.006a			< 0.001a
None	Reference			Reference
Yes	2.21	1.26-3.87		4.69	2.32-9.51
COPD			< 0.001a			0.020a
None	Reference			Reference
Yes	6.26	2.32-16.88		4.82	1.28-18.18
Neoadjuvant chemotherapy
None	Reference
Chemotherapy	0.73	0.33-1.59	0.426
Immunochemotherapy	0.60	0.32-1.13	0.113
Neoadjuvant radiotherapy			0.057			0.051
None	Reference			Reference
Yes	2.75	0.97-7.75		3.69	1.00-13.71
Tumor size	0.94	0.83-1.07	0.352
FEV1	0.13	0.07-0.23	< 0.001a	0.13	0.06-0.27	< 0.001a
Ca	0.04	0.00-0.77	0.033a	0.06	0.00-2.39	0.134
Hemoglobin	0.98	0.96-0.99	0.002a	0.99	0.97-1.00	0.119
Albumin	0.82	0.76-0.88	< 0.001a	0.83	0.76-0.90	< 0.001a

^aP < 0.05.

COPD: Chronic obstructive pulmonary disease; FEV1: Forced expiratory volume in the first second; Ca: Calcium; OR: Odds ratio; CI: Confidence interval.

Table 4 Checklist of stratification model

Variable	Parameter	Condition
Age	0.026	Continuous variables
History of alcohol consumption	1.546	Positive
COPD	1.573	Positive
Neoadjuvant radiotherapy	1.307	Positive
FEV1	-2.044	Continuous variables
Ca	- 2.824	Continuous variables
Hemoglobin	-0.013	Continuous variables
Albumin	-0.190	Continuous variables

COPD: Chronic obstructive pulmonary disease; FEV1: Forced expiratory volume in the first second; Ca: Calcium.