Wang et al reported a nomogram for predicting the risk of hospital-acquired infection for a pediatric patient with spinal cord injury. This can help to identify a patient who may benefit from preventative measures. The authors are from Nanjing University, Hangzhou University, and Southeast University in China.
Patient selection: pediatric patient with spinal cord injury, from 3 to 15 years
Parameters:
(1) age of the patient in years
(2) time from injury to the hospital (ITH) in hours, from 0 to 60 hours
(3) history of pulmonary infection
(4) history of urinary tract infection
(5) urine urobilinogen test
(6) damaged spinal cord segment
(7) admission ASIA score
points for age =
= 115.5 - (7.7 * (age))
points for time to hospital =
= 1.282 * (hours)
Parameter
|
Finding
|
Points
|
history of pulmonary infection
|
no
|
0
|
|
yes
|
41.6
|
history of urinary tract infection
|
no
|
0
|
|
yes
|
30.3
|
urine urobilinogen
|
negative
|
0
|
|
positive
|
17
|
damaged spinal cord segment
|
sacrococcygeal
|
0
|
|
lumbar
|
16
|
|
thoracic
|
32
|
|
cervical
|
48
|
admission ASIA score
|
A
|
54.8
|
|
B
|
41.1
|
|
C
|
27.4
|
|
D
|
13.7
|
|
E
|
0
|
total score =
= SUM(points for all of the parameters)
Interpretation:
• minimum score: 0
• maximum score: 361.1
• The higher the score the greater the risk of a hospital-acquired infection.
Score
|
Value of X
|
66.8 to 137.4
|
(0.000135 * ((score)^2)) - (0.012 * (score)) - 1.998
|
137.4 to 258.5
|
(-0.0001167 * ((score)^2)) + (0.0804 * (score)) - 9.984
|
probability of HAI =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is reported as 0.96 for the training and 0.92 for the validation cohorts.