Zhang and Yang reported a model for predicting a prolonged pancreaticoduodenectomy. The authors are from Chinese Academy of Medical Sciences & Peking Union Medical College in China.
Patient selection: pancreaticoduodenectomy (Whipple procedure)
Prolonged: > 300 minutes
Parameters:
(1) body mass index (BMI) in kg per sq m
(2) blood type
(3) history of pancreatitis
(4) hypertension
(5) diabetes
(6) preoperative complication
(7) pylorus preservation
(8) blood loss
(9) vascular resection
points for BMI =
= (4.167 * (BMI)) - 58.333
Parameter |
Finding |
Points |
blood type |
1 (B+) |
0 |
|
2 (O+) |
18.7 |
|
3 (A+) |
37.3 |
|
4 (AB+) |
55.9 |
history of pancreatitis |
no |
43.2 |
|
yes |
0 |
hypertension |
no |
0 |
|
yes |
13.3 |
diabetes |
no |
0 |
|
yes |
14.3 |
preop complications |
no |
0 |
|
yes |
18.4 |
pylorus preservation |
no |
51.1 |
|
yes |
0 |
blood loss |
<= 500 mL |
0 |
|
> 500 |
67.3 |
vascular resection |
no |
0 |
|
yes |
20.3 |
where:
• The "+" with the blood type is assumed to reflect status of ABO antigen and not be related to RhD status.
total score =
= SUM(points for all of the parameters)
Interpretation:
• minimum score: 0
• maximum score: 383.8
value of X =
= (0.0231 * (score)) - 4.913
percentage of patients with prolonged surgery =
= 1 / (1 + EXP((-1) * X))
Performance:
• The area under the ROC curve is 0.79.
Specialty: Gastroenterology