Maintenance-related complications and the main causes of unplanned extubation
In 334 placed CVCs, 102 (35.4%) had the complications related to CVC insertion (Table 1). The most frequent complication was catheter-related infection occurring in 44 (13.17%) patients, then catheter malposition of CVCs occurring in 28 (8.38%) patients, puncture site exudate in 16 (4.79%) patients. The main reasons for unplanned removal were catheter-related infection (13.17%), catheter malposition (4.19%), puncture site exudate (4.79%). There were two or more complications of the CVCs in 11 cases, catheter-related infection and malposition in 3 cases, catheter malposition and puncture site exudate in 5 cases, catheter-related infection and catheter occlusion in 1 case. Three kinds of complications were found in 2 cases, including catheter malposition, catheter occlusion and catheter associated infection.
Table 1 The Incidence of complications and unplanned removal of CVC
Complications
|
Total number of complications n (%)
|
Extubation cases in advance n (%)
|
Catheter-related infection
|
44 (13.17%)
|
44 (13.17%)
|
Catheter malposition
|
28 (8.38%)
|
14 (4.19%)
|
Puncture site exudate
|
16 (4.79%)
|
16 (4.79%)
|
Obstruction
|
10 (2.99%)
|
4 (1.20%)
|
Accidental removal
|
2 (0.60%)
|
2 (0.60%)
|
Central venous thrombosis
|
2 (0.60%)
|
2 (0.60%)
|
Combined
|
102 (30.54%)
|
88 (24.55%)
|
Complications Related to Catheter Placement and Duration
Maintenance-related complications except for center vein thrombosis were independent of access sites (jugular vein, femoral vein and subclavian vein) (P > 0.05). Due to the low incidence of accidental removal and central venous thrombosis, the incidence of accidental removal of the jugular vein catheter was compared with that of the femoral vein catheter and subclavian vein catheter by Fisher's test (P > 0.05), which showed no statistical significance.
The longer the duration of catheterization, the higher the incidence of total complications, such as catheter-related infection, puncture site exudate and obstruction. (P < 0.05). The comparison between different indwelling time is statistically significant. When catheters are in place for extended periods, the catheter probably plays a major role in providing access for microorganism to the bloodstream by migrating endoluminally.
Table 2 The incidence of complications at different access sites
Complications
|
Total
(n=334)
n (%)
|
Femoral
(n=58, 17.36%)
n (%)
|
Jugular
(n=257, 76.95%)
n (%)
|
Subclavian
(n=19, 5.69%)
n (%)
|
χ2
|
P
|
Catheter-related infection
|
44 (13.17%)
|
8 (12.12%)
|
32 (12.90%)
|
4 (21.05%)
|
2.435
|
0.271
|
Catheter malposition
|
28 (8.38%)
|
2 (3.03%)
|
24 (9.68%)
|
2 (10.53%)
|
2.995
|
0.289
|
puncture site exudate
|
16 (4.79%)
|
3 (4.55%)
|
13 (5.24%)
|
0
|
3.213
|
0.332
|
Obstruction
|
10 (2.99%)
|
4 (6.06%)
|
6 (2.42%)
|
0
|
5.065
|
0.063
|
Accidental removal
|
2 (0.60%)
|
0
|
2 (0.81%)
|
0
|
-
|
0.465
|
Central venous thrombosis
|
2 (0.60%)
|
2 (3.03%)
|
0
|
0
|
-
|
0.054
|
Combined
|
102 (30.54%)
|
19 (28.79%)
|
77 (31.05%)
|
6(31.58%)
|
0.095
|
0.543
|
Note: "-" is Fisher's test
Table 3 The incidence of complications at different catheterization time
Complications
|
Total
(n=334)
n (%)
|
≤ 5 d
(n=131, 39.22%)
n (%)
|
5 ~ 10 d
(n=141, 42.22%)
n (%)
|
> 10 d
(n=62, 18.56%)
n (%)
|
χ2
|
P
|
Catheter-related infection
|
44 (13.17%)
|
11 (3.29%)
|
21 (6.29%)
|
13 (3.89%)
|
5.783
|
0.021
|
Catheter malposition
|
28 (8.38%)
|
9 (2.69%)
|
14 (4.19%)
|
5 (1.50%)
|
1.874
|
0.071
|
Puncture site exudate
|
16 (4.79%)
|
6 (1.20%)
|
4 (1.20%)
|
8 (2.40%)
|
10.763
|
0.011
|
Obstruction
|
10 (2.99%)
|
0
|
5 (1.50%)
|
5 (1.50%)
|
14.643
|
0.001
|
Accidental removal
|
2 (0.60%)
|
0
|
1 (0.30%)
|
1 (0.3%)
|
-
|
0.123
|
Central venous thrombosis
|
2 (0.60%)
|
2 (0.60%)
|
0
|
0
|
-
|
0.435
|
Combined
|
102 (30.54%)
|
28 (21.37%)
|
45 (31.91%)
|
32 (51.61%)
|
19.764
|
0.000
|
Note: "-" is Fisher's test
Catheter-related infection
In catheter-related infections cases, 28 (8.38%) cases were infections of the access site. Catheter-related bloodstream infections (CRBI) developed in 13 (3.89%) patients, the infection rate was 5.3/1000 catheter days. 3 (0.9%) cases were catheter colonization. 47 pathogens were monitored out 44 cases of catheter-related infections. Details are shown in Table 4. Three patients detected two kinds of pathogenic microorganisms (Staphylococcus epidermidis and Candida albicans). Monitoring results: 22 cases were Staphylococcus epidermidis and Acinetobacter baumannii 6 cases, 7 cases with Klebsiella pneumoniae, Pneumococcal in 2 cases, 2 cases of Staphylococcus aureus, Candida albicans 8 cases.
Analysis of the frequency of infection by insertion site showed no differences between the jugular, femoral and subclavian vein access. The risk of catheter-related infection was higher when the duration of catheterization was longer (P = 0.021).
Table 4 Distribution and composition ratio of multi-drug resistant pathogens
Bacterial infection
|
Number (n)
|
Constituent ratio (%)
|
Staphylococcus epidermidis
|
22
|
46.81%
|
Acinetobacter baumannii
|
6
|
12.77%
|
Klebsiella pneumoniae
|
7
|
14.89%
|
Pneumococcal
|
2
|
4.26%
|
Staphylococcus aureus
|
2
|
4.26%
|
Candida albicans
|
8
|
17.02%
|
Combined
|
47
|
100.00%
|
Results of analysis of risk factors associated with total complications
To evaluate the relationship between the risk of factors and total complications, we studied the probability of risks factors by univariate analysis. Catheter-related complications were not related to the gender, age, weight, catheter brand, access site, place, or whether the procedure was performed by anesthetist or pediatrician and so on. Univariate analysis showed suture-off, catheter type, duration of catheterization, bleeding of insertion site, the patients’ maximun channels of intravenous infusion were statistically significant (P<0.05). The five risk factors were possible influencing factors for the complications of CVCs in PICU (Table 5).
The five risk factors with statistically significant in univariate analysis were as independent variables, and then a multivariate logistic regression analysis model was constructed based on the risk factors. Multivariate Logistic regression analysis showed suture-off, bleeding of insertion site, and the maximum channels of intravenous infusion channels of patients, were three dominant risk factors for CVCs-related complications.
Table 5 Univariate anaiysis of factors affecting CVC complications
Factors
|
Kinds
|
No complications (cases)
|
Complications (cases)
|
χ2
|
P
|
Gender
|
male
|
116
|
55
|
2.893
|
0.193
|
female
|
114
|
49
|
Age
|
≤ 3
|
63
|
27
|
0.432
|
0.563
|
> 3
|
169
|
75
|
The puncture site person
|
anesthetist
|
169
|
76
|
0.004
|
0.843
|
pediatrician
|
63
|
26
|
The puncture place
|
operating room
|
144
|
65
|
0.432
|
0.321
|
PICU
|
88
|
37
|
Suture-off
|
yes
|
9
|
27
|
11.432
|
0.000
|
no
|
223
|
75
|
Catheter type
|
double cavities
|
34
|
7
|
8.732
|
0.002
|
three cavities
|
198
|
95
|
Duration of catheterization
|
≤ 5 d
|
102
|
31
|
9.821
|
0.01
|
5~ 10 d
|
94
|
43
|
> 10 d
|
36
|
28
|
Parenteral nutrition
|
yes
|
121
|
54
|
0.043
|
0.982
|
no
|
111
|
48
|
Catheter brand
|
Yixinda
|
161
|
67
|
1.232
|
0.223
|
Di’ao
|
71
|
35
|
Access site
|
Femoral vein
|
39
|
19
|
0.043
|
0.964
|
Jugular vein
|
180
|
77
|
Subclavian vein
|
13
|
6
|
Bleeding of insertion site
|
yes
|
85
|
62
|
11.213
|
0.000
|
no
|
147
|
40
|
Maximun channels of intravenous infusion
|
≤ 4
|
167
|
38
|
12.229
|
0.000
|
> 5
|
65
|
64
|
Table 6 Multivariate unconditional Logistic regression analysis of complication
Factors
|
β
|
χ2
|
OR
|
95%CI
|
P
|
Precedence
|
Suture-off
|
1.632
|
16.923
|
4.858
|
2.313 ~ 10.321
|
0.001
|
1
|
Bleeding of insertion site
|
0.701
|
7.203
|
1.837
|
1.132 ~ 3.041
|
0.008
|
2
|
Maximun channels of intravenous infusion
|
0.185
|
5.921
|
1.238
|
1.032 ~ 1.543
|
0.009
|
3
|
Catheter type
|
-0.053
|
0.321
|
0.985
|
0.987 ~ 1.255
|
0.168
|
4
|
Duration of catheterization
|
-0.051
|
0.289
|
0.945
|
0.778 ~ 1.219
|
0.184
|
5
|